Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reducing mitral valve regurgitation through constriction of the annulus Holloway, Daniel Douglas 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-0736.pdf [ 9.41MB ]
Metadata
JSON: 831-1.0092755.json
JSON-LD: 831-1.0092755-ld.json
RDF/XML (Pretty): 831-1.0092755-rdf.xml
RDF/JSON: 831-1.0092755-rdf.json
Turtle: 831-1.0092755-turtle.txt
N-Triples: 831-1.0092755-rdf-ntriples.txt
Original Record: 831-1.0092755-source.json
Full Text
831-1.0092755-fulltext.txt
Citation
831-1.0092755.ris

Full Text

REDUCING M I T R A L V A L V E REGURGITATION T H R O U G H CONSTRICTION OF T H E ANNULUS by DANIEL DOUGLAS H O L L O W A Y B.Sc, the University of British Columbia, 2001 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES (EXPERIMENTAL MEDICINE) T H E UNIVERSITY O F BRITISH C O L U M B I A April 2006 © Daniel Douglas Holloway, 2006 Abstract Introduction Mitral valve regurgitation ( M R ) is a result of hemodynamic incompetence of the mitral valve. A n in-vitro model of M R was developed to investigate type II and type III-b mitral dysfunction. The role of annular constriction in the reduction of M R was investigated. Methods A porcine heart was mounted in the apparatus. Chords were shortened, lengthened, and cut to create a restricted leaflet, a prolapsed leaflet, and a flail leaflet, respectively. A novel annular constriction device was implanted onto the outside surface of the heart. The magnitude of M R was measured for each created lesion at annular constrictions of 10%, 30%, and 50%. The regurgitation flow rate and the change in regurgitation relative to the disease regurgitation rate were calculated. Results Creation of mitral valve lesions produced M R significantly greater than measured baseline values (pO.OOOl). Annular constriction alone significantly reduced M R for a flail leaflet (p=0.0054), prolapsed leaflet (p=0.0018), and restricted leaflet (p=0.0011). The effect of lesion type does not have a significant difference on the regurgitation reduction for 10%, 30%, or 50% annular constriction. The initial disease severity appears to affect the regurgitation flow rate at 10%) and 30% but not 50% annular constriction. Discussion Annular constriction, without any other surgical alteration to the valve, was found to effectively reduce M R for type Il l-b and type II mitral dysfunction. M R is reduced through improved leaflet coaptation as well as through formation of a monoleaflet valve where the posterior leaflet is functionally nullified and therefore lesions associated with the posterior leaflet become irrelevant. The mechanisms of M R reduction apply similarly to restricted, prolapsed, and flail leaflets. For this reason the three lesions react similarly to annular constriction. The regurgitation flow rate is dependent on the initial disease severity i f the annulus is constricted 10% or 30%. If the annulus is constricted 50%, the regurgitation flow rate is reduced to the approximate baseline value regardless of the initial disease severity. Conclusion Annular constriction alone without any other surgical repair to the valve produced significant reductions in M R for type Il l-b restricted leaflet and type II prolapsed and flail leaflets. i i Table of Contents Abstract 1 1 Table of Contents i i i List o f Tables v i i List o f Figures v i i i Acknowledgments x Chapter 1 - Introduction 1 1.1 Purpose 1 1.2 What is Mitral Regurgitation? 1 1.3 Anatomy of the Heart 2 1.4 Etiology o f Mitral Regurgitation 4 1.5 Types of Mitral Valve Dysfunction '. 5 1.5.1 Type I Dysfunction Causing Mitral Regurgitation 5 1.5.2 Type II Dysfunction Causing Mitral Regurgitation 5 1.5.3 Type III Dysfunction Causing Mitral Regurgitation 7 1.6 Medical Management of Mitral Regurgitation 9 1.7 Surgical Therapies for Mitral Regurgitation 10 1.7.1 Mitral Valve Replacement or Repair 10 1.7.2 Annuloplasty 11 1.7.3 Leaflet Resection 12 1.7.4 Edge-to-Edge Alf ier i Repair 13 1.7.5 Correction of Ventricular and Subvalvular Deformations 14 1.7.6 Experimental Manipulation of Papillary Muscles 14 1.8 Minimal ly Invasive Repairs 15 1.8.1 Alf ier i Repair 15 1.8.2 Coapsys ® Device Annuloplasty 16 1.8.3 Percutaneous Device Annuloplasty 16 1.9 Which Aspects of Mitral Valve Repair are Most Effective? 17 Chapter 2 - Methods 19 2.1 Overview o f Established Models of Mitral Regurgitation 19 2.2 Overview o f the Model of Mitral Regurgitation Used in this Study 20 2.3 Design of the Heart Cage 20 2.4 Heart Preparation 24 2.5 Mounting the Heart 24 2.6 Annulus Markers 29 2.7 Device and Module design 31 2.7.1 Design Criteria 31 2.7.2 Device Design Overview 32 2.7.3 Design o f the Standard Modules 33 i i i 2.7.4 Modules with Alternative Footprints 34 2.7.5 Construction o f the Device 35 2.7.6 Implantation of the Modules 36 2.7.7 Module Location across the Annulus 36 2.7.8 Tensioning Suture Implantation 36 2.8 The Model of Mitral Valve Regurgitation 37 2.8.1 Creation of Type II and Type Il l-b Mitral Valve Dysfunction 37 2.8.2 Creation of the Restricted Posterior Leaflet 37 2.8.3 Creation of a Prolapsed Leaflet 39 2.8.4 Creation of Ruptured Chords 40 2.9 Experimental Protocol 40 2.9.1 Overview , 40 2.9.2 Measurement o f Annulus Markers 40 2.9.3 Calculation of the Annulus Area 41 2.9.4 Measurement of Disease Regurgitation F low Rate 41 2.9.5 Qualitative Assessment of Initial Disease Regurgitation Flow Rate 42 2.10 Statistical Analysis 42 Chapter 3 - Results 43 3.1 Analysis of the Mitral Regurgitation Model 43 3.1.1 Changes in the Chord Length , 43 3.1.2 Changes in the Anterior-Posterior (A-P) Distance 43 3.1.3 Changes in the Annulus Area 44 3.1.4 Changes in the Annulus Circumference 45 3.1.5 Measured Changes in the Distances between Annulus Markers 46 3.1.6 Regurgitation Flow Rate for the Baseline and created Lesions 48 3.1.7 Games-Howell test of the Regurgitation F low Rate for the Baseline and 5 Q Created Lesions 3.2 The Effect o f Annular Constriction on Mitral Regurgitation 50 3.2.1 The Effect of Annular Constriction on a Fla i l Leaflet due to Cut Chords 50 3.2.2 Games-Howell test of the Effect o f Annular Constriction on a Fla i l Leaflet. 52 3.2.3 The Effect o f Annular Constriction on a Prolapsed Leaflet due to ^ Lengthened Chords 3.2.4 Games-Howell test of the Effect o f Annular Constriction on a Prolapsed ^ Leaflet 3.2.5 The Effect o f Annular Constriction on a Restricted Leaflet due to ^ shortened Chords 3.2.6 Games-Howell test of the Effect of Annular Constriction on a Restricted ^ Leaflet 3.3 The Effect o f the Lesion Type 58 3.3.1 The Effect o f Lesion Type on the Abil i ty to Reduce Mitral Regurgitation at ^ 10% Annular Constriction iv 3.3.2 The Effect of Lesion Type on the Abi l i ty to Reduce Mitral Regurgitation at ^ 30% Annular Constriction 3.3.3 The Effect of Lesion Type on the Abi l i ty to Reduce Mitral Regurgitation at ^ 50% Annular Constriction 3.4 The Effect of the Initial Disease Severity 62 3.4.1 The Effect of the Initial Disease Severity on the ability to Reduce ^ Regurgitation through Annular Constriction 3.4.2 Welch A N O V A Testing of the Effect of Disease Severity 64 3.5 Qualitative Observations of the Mitral Valve during Annular Constriction 65 Chapter 4 - Discussion 66 4.1 Overview 66 4.2 The Reduction of Mitral Regurgitation through Annular Constriction 66 4.2.1 Hypothesis of Annular Constriction 66 4.2.2 Analysis of the Regurgitation Flow Rates 66 4.3 The Effect of Annular Constriction on Type Il l-b Dysfunction 67 4.3.1 Evidence Supporting the Importance of Annular Constriction For Type III- ^ b Dysfunction 4.3.2 Annular Constriction Using the Coapsys Device 68 4.3.3 Annular Constriction Using Undersized Rings 69 4.4 Evidence Supporting the Importance of Annular Constriction For Type II ^ Dysfunction 4.5 Why Does Annular Constriction Reduce Mitral Regurgitation? 69 4.5.1 Improvement of Leaflet Coaptation 69 4.5.2 Formation of a Monoleaflet Mitral Valve 70 4.6 Possible Changes in the Chordal Force Distribution has an Unknown Effect ? » on M R 4.7 The Effect of Lesion Type 73 4.7.1 The Effect of Lesion Type on the Abi l i ty to Reduce Mitral Regurgitation ^ through Annular Constriction 4.7.2 Limitations of the Calculated Regurgitation Reduction Percentage 74 4.8 The Effect of Disease Severity on the Abi l i ty to Reduce Mitral Regurgitation ^ Through Annular Constriction 4.9 The Ideal Magnitude of Annular Constriction 76 4.10 The Creation of Regurgitation through Valvular Lesions 76 4.11 Advantages of the Novel Model of Mitral Valve Disease. 77 4.11.1 Real Hearts are used for the Creation of Mitral Regurgitation 77 4.11.2 Visibi l i ty of the Mitral Valve 78 4.11.3 Accessibility of the Mitral Valve 78 4.12 Limitations o f this Mitral Valve Disease Model 79 4.12.1 The Heart is Static and Lacks Hemodynamically Accurate Fluid F low 79 4.12.2 Comparison with Clinically Measured Mitral Regurgitation 79 v 4.12.3 Absence of Atrial Pressure 80 4.12.4 Healthy Hearts are used to Simulate Disease 80 4.13 Annuloplasty Using the Novel Device 81 4.13.1 Analysis of Changes in the Anterior-Posterior Distance 81 4.13.2 Analysis of the Annulus Area 81 4.13.3 Comparison of Annulus Area Reduction 81 4.13.4 Stenosis after Annuloplasty. 82 4.13.5 Analysis of Annulus Circumference 82 4.14 Conformational Changes in Sections along the Annulus 83 4.14.1 Changes in the Distances between Markers along the Anterior Annulus. . . . 83 4.14.2 Changes in the Distances between Markers along the Posterior Annulus. . . 83 4.14.3 Changes in the Distances between Markers Spanning the Valve Orifice. . . . 84 Chapter 5 - Conclusion 85 5.1 Annular Constriction Reduces Mitral Regurgitation 85 5.2 The Lesion Type does not affect the Reduction of Regurgitation due to g<-Annular Constriction 5.3 The Initial Disease Severity Affects the Regurgitation Rate at 10% and 30% ^ but not 50% Annular Constriction 5.4 Alteration of the Chords Created Fla i l , Prolapsed, and Restricted Posterior ^ Leaflets 5.5 Avenues of Future Study 86 References 87 v i List of Tables Table 3 . 1 - The average change in chord length after chord shortening and chord ^ lengthening procedures Table 3.2 - Games-Howell multiple comparison test of regurgitation flow rates for the normal baseline, flail leaflet due to cut chords, prolapsed leaflet due to 50 lengthened chords, and restricted leaflet due to shortened chords Table 3.3 - Games-Howell multiple comparison test of regurgitation flow rates for ^ 0% (disease), 10%, 30%, and 50% annular constriction for hearts with cut chords. Table 3.4 - Games-Howell multiple comparison test of regurgitation flow rates for 0% (disease), 10%, 30%, and 50% annular constriction for hearts with lengthened 54 chords Table 3.5 - Games-Howell multiple comparison test of regurgitation flow rates for 0% (disease), 10%, 30%, and 50% annular constriction for hearts with restricted 57 leaflets due to shortened chords Table 3.6 - Welch A N O V A tests of the effect of initial disease severity on the ^ ability to reduce M R through annular constriction of 10%, 30%, and 50% vi i List of Figures Figure 1 .1- Geometric relations of the left ventricle with the mitral and aortic ^ valves Figure 1.2 - The coapting anterior and posterior leaflets of the mitral valve as ^ viewed with the left atrium removed Figure 1 .3- Type II dysfunction: a prolapsed posterior leaflet due to elongated ^ chords Figure 1.4 - Type II dysfunction: a flail posterior leaflet due to ruptured ^ chords Figure 1.5- Type Il l-b restricted posterior leaflet due to increased tethering ^ forces Figure 2.1 - A general overview of the heart cage design 22 Figure 2.2 - A picture of the actual heart cage. 23 Figure 2.4 - A top view of the heart mounted in the heart cage 26 Figure 2.5 - A view of the right ventricle with the mounting rods attached 27 Figure 2.6 - The view of the left ventricle when the heart is mounted in the 2g heart cage Figure 2.7 - The spikes are placed through the ventricle muscle wall and 29 fastened onto rods with various mounting angles Figure 2.8 - The location and number of markers placed along the annulus 30 Figure 2.9 - A picture of the exposed mitral valve and annulus with markers in place. Also pictured are the device modules implanted on the outside surface 31 of the heart Figure 2 . 1 0 - The device modules implanted on the outside surface of the ^3 heart Figure 2.11 - The side, bottom, and top views of the standard module design... 34 Figure 2 . 1 2 - Attitude adjustment of the modules is performed though ^ tensioning o f the top and bottom transverse sutures Figure 2 . 1 3 - Footprints of the standard and alternative modules 35 Figure 2.14 - Artificial chords were created using 2-0 sutures anchored to the papillary muscle Figure 3.1 - Average measured distance across the mitral valve (A-P distance) ^ (mm) with standard error for the baseline and constricted annulus Figure 3.2 - Average calculated annulus area (mm 2) with standard error for the ^ baseline and constricted annulus Figure 3.3 - Average measured annulus circumference (mm) with standard ^ error for the baseline and constricted annulus Figure 3.4 - Average change in distance between markers along the annulus ^ for annular constrictions of 10%, 30%), and 50%) v i i i Figure 3.5 - The regurgitation flow rate (ml/min) for the baseline heart and the flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and ^ restricted leaflet due to shortened chords. A Welch A N O V A test produces p O . O O O l . . Figure 3.6 - Regurgitation flow rates (ml/min) for the baseline and cut chords with annular constriction of 0% (disease), 10%, 30%, and 50%). Welch _. A N O V A test of 0% (disease), 10%, 30%, and 50% annular constriction produces p = 0.0054 Figure 3.7 - Regurgitation flow rates (ml/min) for the baseline and lengthened chords with annular constriction of 0% (disease), 10%), 30%, and 50%. Welch ^ A N O V A test of 0% (disease), 10%, 30%, and 50% annular constriction produces p = 0.0018 Figure 3.8 - Regurgitation flow rates (ml/min) for the baseline and shortened chords with annular constriction of 0% (disease), 10%, 30%, and 50%. Welch _ , A N O V A test of 0% (disease), 10%, 30%, and 50% annular constriction produces p = 0.0011 Figure 3.9 - Regurgitation reduction (%>) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to ^ shortened chords at 10% annular constriction. Analysis of variance test produces p = 0.1537 Figure 3.10 - Regurgitation reduction (%) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to ^ shortened chords at 30% annular constriction. Analysis of variance test produces p = 0.5527 Figure 3.11 - Regurgitation reduction (%) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to ^ shortened chords at 50% annular constriction. Welch A N O V A test produces p = 0.2427 Figure 3.12 - Average measured regurgitation flow (ml/min) for hearts with mild, moderate, or severe mitral regurgitation before annular constriction. 64 Annular constrictions of 0% (Disease), 10%, 30%, and 50% are represented.... ix Acknowledgments I would like to thank everybody who made this study possible. A special thanks to my supervisors Dr. Peter Skarsgard and Dr. Sam Lichtenstein. I appreciate the opportunity I was given to study such an interesting and exciting topic. Your combined advice and guidance has been exemplary. I am thankful for the great help from my review committee: Dr. Richard Cook, Dr. Keith Walley, and Dr. Jian Ye . I appreciate the time you have taken to help further develop this study. Your insight is invaluable. A t the Experimental Medicine office, the Director Dr. Vincent Duronio and the Graduate Secretary Patrick Carew have been very understanding and helpful. I appreciate the work on my behalf. Also thanks to Dr. Ji Zhang who first exposed me to the wi ld world of cardiac surgery. I consider myself fortunate to have been able to seek your advice over the past several years. Kr is Gillespie at the Jack Be l l Research Centre has been tremendously helpful. I appreciate the consideration I received whether it was through receiving hearts or receiving advice. I am very thankful to be supported by an amazing network of friends and family. Constant encouragement was always present, and help was always available. These intangibles are of utmost value. Thank-you M o m , Dad, Karen, Grandma; and of course, Melissa. Chapter 1 - Introduction 1.1 Purpose Mitral valve regurgitation is a very complex and potentially fatal disease. Decades of research and experimentation have enabled physicians today to administer life saving therapies. The study performed is an investigation of mitral valve regurgitation reduction through surgical constriction of the mitral annulus. 1.2 What is Mitral Regurgitation? Mitral regurgitation ( M R ) is the flow of blood in the reverse direction during systole. Simply put, it is a leaky valve in the heart. Regurgitant blood flows from the left ventricle, past the mitral valve into the left atrium. The mitral valve, also known as the bicuspid valve, separates the left atrium from the left ventricle [1]. The valve opens to let blood flow from the atrium into the ventricle. A normal competent mitral valve closes when the ventricle contracts; this prevents blood from flowing back into the atrium. When the mitral valve is not competent, ventricular contraction results in some of the blood being pumped back from the ventricle into the left atrium and pulmonary venous system [2]. Retrograde blood flow back into the atrium can be detrimental because it creates elevated pressure and volume in the left atrium [3]. In conjunction, since some of the blood is pumped back into the left atrium and pulmonary system, less blood is pumped to the rest of the body via the systemic system [3]. Therefore, in order to pump enough blood to supply the needs of the systemic system, the heart must increase the volume of blood pumped for each cardiac cycle [2]. Increases in the left ventricular stroke volume can eventually lead to a volume overload [3]. The left ventricle becomes dilated to the point that the contractility of the ventricle muscle walls decrease [2]. The heart reaches a point where it is no longer capable of pumping the required amount of blood through the systemic system [4]. Also , due to an elevation in left atrial pressure and therefore post capillary pressure, pulmonary congestion occurs. This leads to shortness of breath. Without medical correction, mitral regurgitation can lead to heart failure and death [3]. 1 1.3 Anatomy of the Heart Although the basic premise of heart function is simple (it pumps the blood throughout the body) the structure of the heart is amazingly complex. For this study the primary concern is the left side of the heart. The pulmonary veins convey oxygenated blood from the lungs to the left atrium. The blood then passes from the left atrium through the mitral valve into the left ventricle. From the left ventricle the blood is pumped past the aortic valve into the aorta. Arteries branching off the aorta bring oxygenated blood to the coronary system, brain, and all the other muscles and tissues throughout the systemic circulatory system. A n illustrated cross-section of the left ventricle and mitral valve apparatus is depicted in Figure 1.1. The mitral valve contains two collagenous leaflets; the anterior leaflet and the posterior leaflet [1]. The leaflets span the valve orifice and are configured so that when functioning properly, the leaflets separate to allow blood flow into the ventricle during diastole (ventricle relaxation) and coapt during systole (ventricle contraction) to prevent retrograde blood flow from the ventricle to the atrium [5]. Figure 1.2 is a picture of the mitral valve leaflets viewed with the left atrium removed. The leaflets are composed of three scalloped subsections, commonly referred to as A l , A 2 , and A 3 for the anterior leaflet and P I , P2, and P3 for the posterior leaflet. The valve orifice is a fibrous ring named the mitral annulus [1]. The annulus is composed o f fibrous connective tissue as well as contractile myocardium [5]. The leaflets are suspended across the annulus as part of a complex structure known as the mitral valve apparatus. The mitral valve apparatus contains the leaflets, chordae tendinae, papillary muscles, and annulus. The leaflets are held in place through their attachment to the annulus as well as through the attachment of many chordae tendinae (chords) [5]. The very strong fibrous chords suspend the free edges o f the leaflets [1]. Their effect is analogous to the ropes of a parachute. When the leaflets coapt during systole the chords resist the force exerted by the pressurized blood and hold the leaflets in place. If the chords were absent the blood would simply push the leaflets past the plane of the annulus and into the atrium. This would produce a leaky valve. 2 The papillary muscles are also an important component of the mitral valve apparatus. The papillary muscles are a pair of myocardial appendages protruding from the ventricular muscle wall into the ventricular lumen [1]. They serve as attachment points for many o f the chords suspending the leaflets [5]. Figure 1 .1- Geometric relations of the left ventricle with the mitral and aortic valves. 3 Figure 1.2 — The coapting anterior and posterior leaflets o f the mitral valve as viewed with the left atrium removed. 1.4 Etiology of Mitral Regurgitation Mitral regurgitation can be the result o f several different types o f disease. The most common causes of M R are myxomatous degeneration, ischemic heart disease, infective endocarditis, idiopathic ruptured chords, rheumatic heart disease, calcification of the mitral annulus, and hypertrophic cardiomyopathy [2, 3, 6, 7]. Less common causes of M R are congenital defects, Marfan's syndrome, and trauma [4]. Worldwide, rheumatic heart disease is the most common cause of M R with an estimated incidence of 35-40% [8]. This reflects the prevalence of rheumatic disease in developing countries where rheumatic fever is still common. In North America, rheumatic fever is rare, and therefore M R due to rheumatic disease is less common [2]. 4 1.5 Types of Mitral Valve Dysfunction 1.5.1 Type I Dysfunction Causing Mitral Regurgitation Carpentier describes three types of dysfunction which lead to mitral valve regurgitation [9]. A regurgitant valve with leaflets capable o f normal motion is said to have type I dysfunction [9]. For type I dysfunction, regurgitation is a result of lesions which do not affect leaflet motion. Lesions such as a perforated leaflet or dilation of the annulus can produce type I regurgitation. In the case of annular dilation, the anterior and posterior leaflets may have full range of motion, but due to the increased distance across the annulus it is not possible for the leaflets to coapt and therefore the valve leaks [10]. 1.5.2 Type II Dysfunction Causing Mitral Regurgitation Type II dysfunction is characterized by increased leaflet motion, often termed leaflet prolapse [9]. Prolapse is said to occur when a segment of a leaflet is located above the plane of the valvular orifice during systole [11]. This is illustrated in Figure 1.3. This valvular disease is one of the most prevalent cardiac abnormalities [12]. A n estimated that 15 mil l ion Americans have a prolapsed leaflet [13]. The commonly accepted prevalence is 2-3 % [14]. The lesions which cause mitral valve prolapse can vary; however, myxomatous degeneration is the most frequent cause of leaflet prolapse [8]. Myxomatous mitral valves have altered mechanical characteristics due to degeneration of the chords and leaflets. M y x o i d chords are larger and heavier than normal chords but they have reduced mechanical strength [15]. The measured breaking strength of myxoid chords are only half that of normal chords [15]. M y x o i d chords are therefore prone to rupture. Myxomatous mitral valve leaflets were found to be more extensible and less stiff than normal leaflets [16]. Myxomatous degeneration also causes thickening and redundancy of the mitral leaflets [14]. The net result of myxomatous degeneration can be a prolapsed leaflet due to elongated or ruptured chords. Geometric alteration of the left ventricle and mitral valve apparatus, such as produced during ischemic deformation, can also lead to a prolapsed leaflet causing regurgitation [17]. Through analysis of a porcine model, Nielsen et al determined that acute cardiac ischemia can produce a prolapsed anterior leaflet at the posteromedial commissural side [17]. The prolapse of the leaflet is believed to be a result of papillary 5 muscle contractile dysfunction [17]. The inability of the papillary muscle to contract and shorten causes prolapse o f the segment of the leaflet connected via chords to the papillary muscle. Various connective tissue disorders, such as Marfan's syndrome, can also result in chordal elongation producing M R [14]. However, these diseases are relatively rare. For example, Marfan's syndrome only accounts for approximately one in five hundred cases of mitral valve prolapse [18]. Fla i l mitral leaflet is a type II dysfunction where a portion of the leaflet is severely prolapsed usually due to chordal or papillary muscle rupture [19]. The prolapsed portion of the leaflet exhibits erratic and untethered motion [20]. This is illustrated in Figure 1.4. A flail mitral leaflet has been associated with poor clinical outcome and an increase in sudden death [19, 21]. Figure 1.3 - Type II dysfunction: a prolapsed posterior leaflet due to elongated chords. 6 Figure 1.4 - Type II dysfunction: a flail posterior leaflet due to ruptured chords. 1.5.3 Type III Dysfunction Causing Mitral Regurgitation Carpentier defines type III dysfunction as a restriction of leaflet motion [9]. Coaptation is impaired due to physical constraints which limit the motion of one or both leaflets. Type III dysfunctions can be further classified into two distinct subsets: I l l -a and Il l-b. Type I l l -a dysfunctions are defined as a restriction in the opening of the valve [22]. Type I l l -a dysfunctions are created by lesions such as chordal thickening, leaflet thickening, and commissural fusion, as is commonly seen in rheumatic disease [9, 22]. Type Il l-b dysfunctions are defined as a restriction in the closing of a valve [11]. The type Il l-b leaflets have an impaired ability to coapt during systole [22]. Ischemic deformation of the ventricular walls and papillary muscles are the primary lesions for type Il l-b restricted leaflets [10]. 7 Type Il l-b dysfunction with M R is a clinically important problem, and has been studied clinically and experimentally. Nielsen et al used an in-vitro model to create M R through the manipulation of papillary muscle alignment [23]. This study found that posterolateral dislocation of the posteromedial papillary muscle created a prolapsed anterior leaflet and a restricted posterior leaflet [23]. Posterolateral dislocation was used because it is believed to mimic the pathologic location of the papillary muscles when the left ventricle is chronically dilated [24]. This posterolateral adjustment in the papillary muscle is believed to create a restricted posterior leaflet for two reasons. First, the posterior leaflet is subjected to increased chordal tethering forces [25]. Shengqiu He et al have experimentally determined that "Increased tethering of the mitral leaflets produced delayed movement toward closure and premature opening of the leaflets in late systole as the transmitral pressure decreased" [26]. Second, in-folding of the posterior leaflet decreases the disposed surface area [25]. A s a result of decreased surface area, there is less coapting force to counterbalance the tethering force causing the posterior leaflet to become further misaligned [25]. The net result is that the posterior leaflet develops a compromised ability to coapt with the anterior leaflet [27]. A restricted leaflet is illustrated in Figure 1.5. The valve is incompetent because the increased tethering forces on the posterior leaflet impair coaptation. Clinically, dilation of the left ventricle produces a type Il l-b restricted posterior leaflet [28]. This is because the papillary muscle is displaced away from the posterior leaflet, thereby increasing the tethering force and producing a restricted posterior leaflet [29]. Gorman et al observed this phenomenon using an animal model where an ischemia induced infarction created deformation of the left ventricle, papillary muscles, and mitral annulus [30]. 8 Figure 1.5 — Type Il l-b restricted posterior leaflet due to increased tethering forces. 1.6 Medical Management of Mitral Regurgitation Medical therapy allows physicians to treat the underlying etiology of M R in some cases (myocardial ischemia, for example), and also to treat the symptoms caused by M R . Good medical therapy may prevent or delay the need for patients to undergo mitral valve surgery [31]. There are a wide variety of drugs effectively used to treat the causes and effects of mitral regurgitation. This summary is a brief description of the medical therapies available. M i l d mitral regurgitation is usually asymptomatic for patients. It is possible to detect mitral dysfunctions and the specific lesions before patients feel any symptoms of cardiac disease. Frequently, asymptomatic patients w i l l not require medial therapy; monitoring for worsening of the dysfunction is all that is required [8]. For these asymptomatic patients, it is advised that they take prophylactic antibiotics when they may be subjected to acute bacteremia such as during dental, gynecologic, urologic, or septic 9 abdominal procedures [2, 8]. The antibiotics are required to prevent infective endocarditis [1]. If mitral regurgitation is believed to be the result of infective endocarditis, then immediate treatment with antibiotics is required to stop the progression of the lesion [2]. Due to the retrograde flow of blood from the heart back into the lungs, mitral regurgitation results in pulmonary venous hypertension. Symptoms are fatigue and dyspnea [1]. In order to treat pulmonary venous hypertension, patients can be prescribed diuretics, vasodilators, and reduced salt diets [1,8]. A diet with lower salt content corresponds with lower total body sodium levels. Diuretics reduce pulmonary edema by removing excess sodium and water from the blood [32]. The net effect of reduced sodium and water in the blood is reduced intravascular volume; this reduces pulmonary venous pressure [32]. Vasodilators reduce pulmonary edema through two processes: 1) they reduce venous return to the heart and pulmonary system 2) they reduce cardiac afterload through reduction of the systemic blood pressure [32]. Through the use of vasodilators, less blood returns to the heart, and the blood is easier to pump out of the heart to the body tissues. Less blood remains congested in the pulmonary circulation [1]. Anticoagulation therapy is frequently used when patients have an enlarged left atrium, atrial fibrillation, or both [31]. These conditions are related to an increased likelihood of blood clots forming in the left atrium [8]. Anticoagulants are beneficial because through various mechanisms they prevent blood clots from forming [32]. Atrial clots can embolize, leading to heart attack or stroke [31]. 1.7 Surgical Therapies for Mitral Regurgitation 1.7.1 Mitral Valve Replacement or Repair When mitral regurgitation leads to symptoms or impairment in ventricular function, surgical therapy is generally recommended. The most common surgical therapies for M R are valve replacement using an artificial valve or valve repair. Valve replacement and valve repair procedures are very common therapies for mitral dysfunction. Between the years 1999 and 2000, over 20,000 patients underwent mitral valve surgical procedures in America alone [33]. O f these, 39% underwent valve repair procedures, and the remaining 61% underwent valve replacement procedures [33]. 10 Valve replacement consists of implantation of an artificial mechanical or tissue valve in place of the diseased mitral valve [34]. There are many different valve designs used routinely for valve replacement procedures. Pioneered by Dr. Ala in Carpentier, mitral valve repair is based on restoring competent function to the valve. Carpentier states: "surgeons are not basically concerned with lesions. We care more about function" [9]. There are many different surgical techniques for mitral valve repair. Surgical manipulation of the ventricle, papillary muscles, annulus, and leaflets are commonly practiced surgical procedures for the reduction or elimination of M R [9, 35, 36]. The general concepts of the most common mitral valve repair techniques w i l l be discussed. 1.7.2 Annuloplasty Annuloplasty is the surgical manipulation of the valve annulus. The goal of annuloplasty is to alter the shape and size of the annulus to a conformation which matches the available leaflet tissue; therefore restoring competency to the valve [37]. In almost all mitral valve repair procedures, an annuloplasty is performed with implantation of a prosthetic ring around the annulus of the mitral valve [9]. The ring provides structural support as well as acting to control the shape and size of the annulus [9]. The basic design of the annuloplasty ring was developed in 1969 by Carpentier [38]. Today, there are several different ring designs which are routinely used. Some rings are completely rigid while others have varying degrees of flexibility [37]. Some rings are implanted across the entire annular circumference; other rings are "incomplete" and only attach to the posterior section of the annulus [37]. Each design offers advantages and disadvantages. The ideal ring design is contested and continually investigated by the medical research community. Under full cardiopulmonary bypass, the prosthetic ring is implanted in an open-heart procedure [37]. Through an incision in the left atrium, the ring is placed onto the annulus directly over the mitral valve orifice [9]. A series of sutures permanently fixes the ring and annulus together. Through this implantation, the annulus is remodeled to take the shape and size of the prosthetic ring [39]. The purpose of ring implantation is to reduce the size of the annulus and to return the shape of the annulus to its natural conformation [9]. Annular lesions lead to asymmetric dilation of the annulus creating 11 dysfunction; annuloplasty corrects the dilation and restores competent function of the valve [37]. Although ring annuloplasty is a widely used and generally effective therapy, the outcome can be unpredictable. For cases o f ischemic M R , ring annuloplasty has been associated with poor long term outcomes [40]. Matsungaga claims that "although immediate results are excellent, some patients develop recurrent M R at mid-term follow up" [41]. A clinical study of patients receiving prosthetic ring implantation and coronary artery bypass grafts for the treatment of ischemic M R , found that one year after surgery, 90% of the patients had only mi ld or no M R [42]. Another clinical study found that three years after mitral valve repair surgery to repair ischemic M R , 71% of patients had mild or no M R [40]. 1.7.3 Leaflet Resection Mitral dysfunction can occur due to degeneration of the valvular leaflets and chords [43]. The ventricular muscle, papillary muscles, and annulus may be perfectly normal but leaflet coaptation is impaired due to prolapsed caused by chord rupture or elongation. When degeneration causes prolapse of the posterior leaflet, surgical repair of the leaflets is a common procedure [36]. The procedures for degenerative M R primarily focus on restoring the natural conformations of the leaflets. The most common repair is based on the procedure developed by Carpentier where a quadrangular resection of the posterior leaflet is performed followed by annular plication and implantation of an annular ring [9]. The middle portion of the posterior leaflet (P2) is removed; the left and right portions of the posterior leaflet (PI and P3) are brought into apposition with leaflet stitches and plication o f the annulus [44]. A sliding valvuloplasty is commonly performed to ensure uniform tension of the leaflets across the plicated annulus [45]. In this procedure, the residual portions of the posterior leaflet are completely detached from the posterior annulus. After approximating the cut edges of PI and P3, the reconstituted posterior leaflet is reattached to the posterior annulus, thus minimizing tension on the leaflet suture line [46]. Plication of the annulus can be performed through several related techniques. Mesana et al use equally spaced sutures to plicate the annulus [43]. Pomerantzeff et al use two Teflon pads to apply the plicating force [47]. 12 After the quadrangular resection of the posterior leaflet and plication of the annulus, it is common practice for a prosthetic annular ring to be implanted to support the repair [46]. The durability of the repair is increased when a ring is implanted [36]. Gi l l inov et al found that failure to implant a prosthetic ring increased the risk of late re-operation [48]. The described quadrangular resection with annular plication and ring implantation has proven to be an effective therapy for M R due to posterior leaflet prolapse [49]. For patients with resection of the posterior leaflet, freedom from re-operation after 10 years is 89-99% [48-51]. Five years after leaflet resection surgery, Sakamoto et al found 98% of patients had mild or no M R [49]. It is generally accepted that quadrangular resection with annuloplasty is an effective procedure for the reduction of M R due to type II dysfunction. 1.7.4 Edge-to-Edge Alfieri Repair Another surgical procedure for the correction of prolapsed mitral leaflets is the Alf ier i repair; also known as the edge-to-edge or the double orifice repair [52]. For this procedure, the free edges of the anterior and posterior leaflets are approximated using sutures [53]. Suturing the central edges of the leaflets together creates an orifice on each side of the suture. This technique has been used clinically since 1991 and it is documented to have acceptable short term results for the reduction of M R [52]. After 4.5 years, 88% of the patients who received the Alf ier i repair had mild or no M R [52]. The Alf ier i repair is intended to be applied in conjunction with an annuloplasty. Alf ier i claims that when a concomitant annuloplasty is not performed, then the freedom from operation is significantly lower [53]. In another study, Alf ier i states that "Univariable and multivariable analysis showed the absence of annuloplasty to be the only independent risk factor for reoperation" [52]. The Alf ier i repair for ischemic disease is controversial. Timek et al performed experiments on ischemic ovine subjects to investigate the efficacy of the Alf ier i repair in the absence of annuloplasty [54]. It was determined that "the Alf ier i repair did not prevent acute ischemic M R . Furthermore, it did not alter the annular, subvalvular, and leaflet geometric distortions associated with acute ischemia" [54]. This report indicates that the Alf ier i repair alone would not be an effective therapy for ischemic M R . 13 1.7.5 Correction of Ventricular and Subvalvular Deformations Ischemic M R is a result of alterations to the annulus, papillary muscles, and the ventricle muscle walls; "Ischemic M R is a ventricular disease, not a valvular disease" [55]. Several researchers have expressed the ideology that surgical repair of ischemic M R should focus on correcting the lesions creating the disruption in leaflet coaptation [55, 56]. It could be argued that i f myocardial infarction leads to changes in the location and orientation of the papillary muscles and left ventricle, then surgical therapy should attempt to rectify the anatomy and functionality of these regions. Mi l le r states that an annular ring would be the logical choice for the cases of ischemic M R where the only pathological alteration was annulus dilation [55]. However, in cases with complex regurgitation such as with a type I l l-b restricted leaflet, perhaps with a ventricular infarct, Mi l l e r believes a simple ring annuloplasty may be insufficient [55]. Since it has been experimentally verified that the papillary muscles and ventricle play a significant role in the creation o f M R , surgical procedures dedicated to altering the papillary muscle orientation have been predicted by some to play a large role in the future of cardiac surgery [57]. 1.7.6 Experimental Manipulation of Papillary Muscles There are several published studies relating to the reduction of M R through surgical manipulation of the papillary muscles. Menicanti et al have reported the results of a clinical trial where left ventricular reconstruction and annuloplasty were performed on 254 patients [58]. For this study, the ventricle was opened at the apex with all surgical manipulations performed from the ventricular lumen. Annuloplasty consisted of a running suture conducted from one trigone across the posterior annulus, to the other trigone [58]. In conjunction with the annuloplasty, left ventricular remodeling was performed using a purse string suture applied near the base of the papillary muscles. The net effect was to bring the papillary muscles closer together, restoring their normal geometry. Postoperative mitral regurgitation was reduced to mild or absent in 84% of the cases [58]. Menicanti believes the described procedure is effective because it addresses both annular and ventricular deformation. He states that "mitral annulus repair is perhaps the first mechanism that relieves volume overload, but the improved left ventricular geometry is essential to maintain the efficacy of the repair" [58]. 14 Using a chronic ovine model, Liel-Cohen et al created ischemic M R due to infarction [59]. Plication of the left ventricle effectively corrected the physical deformation created by the infarction. This resulted in a reduction or elimination of M R as a result of correcting the location of the papillary muscles [59]. In this study, it was possible to reduce or eliminate M R without any surgical manipulation of the annulus or valve leaflets. The competency of the valve was restored wholly through alteration of the geometric orientation of the papillary muscles via left ventricular plication [59]. The ability to reduce M R through manipulation of the papillary muscles only, without annular constriction, has also been reported by Hung et al [35]. Using an ovine model of ischemic M R , the papillary muscles were repositioned through the application of a novel device [35]. A Dacron patch-balloon device was sutured onto the epicardial surface of the myocardium over the infarcted areas. Inflation of the balloon created conformational changes in the ventricular muscle wall which created changes in the position and orientation of the papillary muscles [35]. It was possible for the investigators to inflate the balloon until echocardiographic evidence of a reduction in M R occurred. Through the manipulation of papillary muscles in this manner, it was possible to reduce or eliminate M R in all cases [35]. 1.8 Minimally Invasive Mitral Valve Repairs 1.8.1 Alfieri Repair The standard edge-to-edge repair as described by Alf ier i requires the patient to undergo cardiopulmonary bypass [53]. Although the procedure is technically simple, the patient is still subjected to the risks and disadvantages of cardiopulmonary bypass. In an effort to repair the mitral valve without the need for cardiopulmonary bypass, several devices based on the Alf ier i repair have been developed. Using echocardiographic guidance, at least three studies have been able to successfully perform an Alf ier i repair on beating heart porcine subjects [60-62]. Novel techniques and devices are currently under development for the purpose of improving the beating heart Alf ier i repairs [61, 62]. Although these studies were generally successful in performing the operation, the ability to reduce M R was not measured and is still unknown. A s previously mentioned, there is strong evidence indicating that the durability of the Alf ie r i repair suffers greatly 15 in the absence of an annuloplasty [54]. Without annuloplasty, it is unknown i f these minimally invasive Alf ier i procedures w i l l produce satisfactory results. 1.8.2 Coapsys ® Device Annuloplasty Developed by Myocor Inc, the Coapsys device is a minimally invasive annuloplasty device. On a beating heart, a polyethylene chord is implanted across the left ventricle from the anterior to the posterior myocardium walls [63]. The chord is situated directly under the mitral valve. A pair of epicardial pads on the end of the chord apply tension to the region of the posterior annulus and posterior myocardium wall [64]. When the device is tensioned the posterior annulus is pulled anteriorly; M R is reduced [63]. This device has been used successfully in several animal trials. Published reports indicate that this device is capable of reducing acute and chronic M R without creating detrimental hemodynamic side-effects [65]. Clinical trials in India and the United States are currently underway [66]. 1.8.3 Percutaneous Device Annuloplasty Percutaneous mitral valve repair is currently under development through several researchers and biomedical companies [67-70]. Remodeling of the mitral annulus is effected by a device placed into the coronary sinus. Since the coronary sinus is located in the atrio-ventricular groove, it is adjacent to the annulus, and thus it is possible to alter the shape of the annulus through implantation of a device. The greatest advantage of such a procedure is that it is possible to implant the device percutaneously [67]. There is no need for invasive surgery, large incisions, or cardiopulmonary bypass. The development o f effective percutaneous mitral valve repairs is in its infancy, yet many believe that permutations of this technique w i l l be the future of annuloplasty [67]. Long term efficacy and durability has yet to be proven in clinical trials. Although the specific design of each percutaneous coronary sinus device varies, they are all intended to produce the same effect: a constriction of the annulus [68-70]. Early results with coronary sinus devices have shown promise; even in the absence o f any direct surgical repair to the leaflets or subvalvular apparatus, constriction of the annulus can reduce M R [68, 70]. 16 1.9 Which Aspects of Mitral Valve Repair are Most Effective? It is well documented that current mitral valve repair procedures are satisfactorily effective in reducing M R for most patients. What is currently unknown is the actual importance of each component of mitral valve repair. A s discussed previously, surgeons can reduce M R through independent manipulation of the annulus, leaflets, papillary muscles, or ventricle wall [35, 43, 59, 71]. Every component has a role, but it has yet to be elucidated which surgical alterations create the greatest reduction of M R . The majority mitral valve procedures used today incorporate surgical manipulation of more than one component of the mitral valve apparatus. For example, quadrangular leaflet resection is typically performed in conjunction with a ring annuloplasty [46]. Open heart procedures afford the surgeons the ability to easily gain access to and subsequently manipulate the papillary muscles, leaflets, and annulus i f desired. A n undeniable trend in the general field of surgery is an evolution towards minimally invasive procedures. The problem with minimally invasive mitral valve repair procedures is that it has not been established which components of repair should be of utmost priority. Should surgeons repair only the leaflets, the annulus, or the subvalvular apparatus? With less invasive strategies being employed more frequently, surgeons may no longer have the ability to manipulate every component of the mitral valve apparatus in one procedure. O f the minimally invasive mitral valve repair procedures currently under development, they all focus on removing valvular dysfunction though manipulation of only one component of the mitral valve apparatus. The minimally invasive Alf ier i procedures manipulate only the leaflets [60-62]. The Coapsys device manipulates the annulus and a portion of the ventricle wall [63, 64]. Percutaneous devices manipulate only the annulus [67-70]. A series of porcine in-vivo ischemic M R experiments performed at the Stanford School of Medicine has found that for type Il l-b dysfunction, annular constriction is a very important component of mitral valve repair [71-73]. Annular constriction alone reduced or abolished M R for acute and chronic ischemic M R with a restricted leaflet [71-73]. Annular constriction appears to play a significant role in the reduction of M R for type Il l-b dysfunction. The effect of annular constriction alone on type II dysfunction has not been directly studied and is not well understood. 17 The purpose of this study is to investigate i f annular constriction alone is capable of reducing M R for type II and type Hl-b mitral dysfunction. If annular constriction moves the anterior and posterior leaflets into closer proximity, then there may be an improvement in leaflet coaptation and therefore a reduction in M R . Also , i f a valve is incompetent due to a lesion affecting the posterior leaflet, then annular constriction may reduce the effect of the lesion by creating a monoleaflet with a functional anterior leaflet and a functionally nullified posterior leaflet. We hypothesize that for type II and type Il l-b mitral dysfunction associated with the posterior leaflet, constriction of the annulus with no other surgical alteration to the mitral valve apparatus w i l l be sufficient in producing a reduction in mitral valve regurgitation due to an improvement in leaflet coaptation and formation of a monoleaflet valve. 18 Chapter 2 - Methods 2.1 Overview of Established Models of Mitral Regurgitation In order to investigate the importance of annular constriction during mitral valve repair, it was necessary to develop a model of mitral regurgitation. In publication, there are several effective models of mitral regurgitation currently being used by various researchers. Sophisticated in-vitro models which use an artificial ventricle have been developed to study the effects of alterations in the configuration and orientation of the papillary muscles, annulus, and leaflets [24-26, 74-77]. The most sophisticated of these models use computer controlled pulsatile flow in an attempt to accurately model the hemodynamic flow through the heart [24, 26, 76]. A model developed by Georgia Tech Institute of Technology, allows researchers to manipulate the size and orientation of the annulus and papillary muscles while measuring the tension on individual chords through force transducers as well as measuring regurgitation through echocardiography [24, 26, 74]. A computer model of M R developed at the University of Washington uses finite element analysis to determine stress patterns and force distribution across the dynamic mitral valve apparatus [78]. This model allows researchers to experimentally adjust the annulus diameter, tissue thickness, and orientation of the papillary muscles [78-80]. With this model, simulations of the regurgitant valves through the cardiac cycle have been reported to closely resemble clinically observed regurgitation [78-80]. In-vivo models of mitral regurgitation have been developed using porcine and ovine subjects. Simple lesions such as cutting chords on a beating heart has been used to create type II dysfunction with M R [81]. Researchers have created type I M R by cutting circular lesions into leaflets [82]. With induced ventricular pacing of 180-230 beats-per-minute for a duration of approximately 15 days, type Il l-b dysfunction with M R has been created through tachycardia induced cardiomyopathy [83]. Surgical ligation of coronary arteries is used to create M R due to acute or chronic ischemia [84-86]. The most well developed ischemic model of M R has been developed by researchers at Stanford University School of Medicine. This model uses small radiopaque markers surgically implanted along left ventricle, papillary muscles, annulus, and mitral leaflets [71, 85, 87]. Videoflouroscopy before and after ischemia allows 19 researchers to record the location of each marker and measure changes in geometry for the ventricle and mitral valve apparatus [71]. In order to detect and measure the magnitude of M R , transesophageal echocardiography with colour Doppler analysis is performed before and after ischemia [87]. 2.2 Overview of the Model of Mitral Regurgitation Used in this Study In order to create a regurgitant mitral valve, many models work under the basic premise that alterations to the geometric orientation of the mitral valve apparatus can produce changes in the chord tension which can lead to impaired leaflet coaptation [27, 79, 88, 89]. This premise is based on clinical and experimental observations where alterations in the location or orientation of the papillary muscles, ventricular muscle walls, annulus, or leaflets can lead to coaptation failures and therefore a regurgitant valve [24, 26, 75, 76, 87]. The model of M R developed for this study follows the same premise. However, instead of altering the orientation of the papillary muscles, leaflet coaptation was impaired through alteration in the chord lengths. A prolapsed posterior leaflet was created by lengthening chords. A flail posterior leaflet was created by cutting , chords. A restricted posterior leaflet was created by shortening chords. 2.3 Design of the Heart Cage In order to perform the required experiments, it was necessary to design a suitable apparatus through which the experiments could take place. The purpose o f the apparatus is to suspend a porcine heart while allowing surgical manipulation of the mitral valve and mitral annulus. In addition, it was required that the apparatus enable easy access to and viewing o f the mitral valve. Furthermore, the apparatus was required to pressurize the left ventricle and enable quantification of the regurgitation flow rate through the mitral valve. A s part of this research project, a suitable apparatus was designed and built. The first design resembled a bird cage; hence, the initial apparatus and subsequent permutations have taken the name "heart cage". Figure 2.1 is a representation of the heart cage. A picture of the heart cage can also be seen in Figure 2.2. A porcine heart is suspended inside a Lexan® cube. A rubber tube is inserted into the lumen of the aorta. Removal of the aortic valve allows water to flow through the tube into the left ventricle. The rubber tube is connected to a column of 20 water. The height of the water determines the water pressure inside the left ventricle. The column is designed so that it is possible to maintain a constant water height, and therefore constant water pressure throughout the experiment. Water is constantly filling the water column. Excess water overflows over the edges of the column and falls into a large funnel at the bottom where it is collected and removed. Since it is imperative to quantify the mitral regurgitation, the regurgitation flow rate is calculated by measuring the regurgitant volume over time. When the heart is suspended inside the Lexan cube, all the regurgitant volume is captured. A large funnel under the heart directs the regurgitant volume into a graduated cylinder for measuring. The flow of water through the heart cage is controlled through two valves. One valve directly proximal to the aorta controls water flow into the heart. The second valve controls water flow into the water column. The frame of the heart cage is constructed with welded mi ld steel. The Lexan cube is composed of five panels of clear Lexan fitted to a steel framework. The water column consists of a P V C pipe fitted to the rubber tube entering the aorta. Fixed to the steel frame o f the Lexan cube, two small lights with flexible shafts illuminate the mitral valve. The flexible shafts can be adjusted to allow illumination of the left ventricular lumen. This is particularly useful when surgically manipulating the mitral valve apparatus. The height from the left ventricle to the top of the water column is 145 cm. This produces a pressure of 107 rnmHg in the left ventricle. This pressure was selected because it is within the range of normal physiologic blood pressure for pigs [90]. The heart cage was designed to allow quick and easy mounting of the heart. Also , it is possible to accommodate different sizes of hearts without compromising the repeatability of every experiment. One paramount consideration in relation to the mounting procedure was that the left ventricle must not be deformed when suspended in the heart cage. It is very important that the left ventricle maintains its natural conformation because the conformation of the left ventricle and its related mitral valve apparatus are the experimental variables in these experiments. In all cases, the heart was suspended in the heart cage without deformation of the left ventricle or mitral valve apparatus. 21 water Column Heart Cage Not To Scale Lex an Cube Water Column Overflow Measured Regurgitant volume Figure 2.1 - A general overview of the heart cage design 22 Figure 2.2 - A picture of the actual heart cage 23 2.4 Heart Preparation In order to provide access for mounting, the right atrial appendage and pulmonary arteries were excised. In addition, in order to expose the mitral valve, the left atrial appendage and pulmonary veins were excised. The aorta was trimmed to a length proximal to the carotid branches and the aortic arch. These modifications produced a heart with intact left and right ventricles, both atria removed above the valvular annuli, and an aorta extending a few centimeters from the left ventricle. The ventricles were flushed with water to remove any blood clots. The aortic valve was carefully removed with attention to leave the mitral valve apparatus untouched. Removal of the aortic valve allows the ventricle to be pressurized with water flowing retrograde through the aorta. It was necessary to ligate the right coronary artery ( R C A ) and the left coronary artery ( L C A ) , as failure to do so would result in coronary perfusion with the effluent collected and included as mitral regurgitant volume. The R C A and L C A were ligated with 2-0 silk sutures approximately 1 cm distal to the branch off the aorta. This eliminated perfusion through the coronary circulation. 2.5 Mounting the Heart The heart is suspended through an arrangement of six adjustable rods extending from the steel frame of the heart cage. The rods can be adjusted in height and length before being locked in place. A l l the rods are attached to the right ventricle. Three rods attach near the apex of the heart; the other three attach near the tricuspid valvular annulus. Figures 2.4, 2.5, and 2.6 are pictures which display the general orientation of the six rods on the right ventricle. Attachment of the rods to the right ventricle occurs through small threaded spikes which travel through the right ventricular muscle wall and fasten to the rods with small threaded nuts. The spikes were placed perpendicularly through the muscle wall . In order to accommodate the curving shape of the heart, especially near the apex; many rods were constructed with varying angles in which to receive the spike. This allowed each heart to be mounted with a selection o f rods which best suited the heart's shape. Figure 2.7 is a drawing of a spike placed through the right ventricle wall and fastened to a mounting rod. The purpose for attaching all the rods to the right ventricle is that this procedure leaves the left ventricle untouched. The left ventricle maintains its natural conformation 24 removing any possibility of the mounting procedure affecting any data when experimental manipulation of the mitral annulus occurs. A s part of the water column, a rubber tube with an outside diameter of % inch is inserted into the aortic trunk. The aorta is fastened over the tube with a locking nylon strap (zap strap). The strap is firmly cinched, creating a watertight seal. A steel support rod attaches to the tubing directly above the aorta. This rod is adjustable along the x, y, and z axes. Any possible orientation of the tubing can be locked in place. It is important to have the tubing adjustable to any orientation because the aorta must retain its normal conformation. 25 Figure 2 . 4 - A top view of the heart mounted in the heart cage. 26 Figure 2.5 - A view o f the right ventricle with the mounting rods attached. 27 28 Right Ventricle Lumen / / Mounting Rod ]Y Spike Figure 2.7 - The spikes are placed through the ventricle muscle wall and fastened onto rods with various mounting angles. 2.6 Annulus M a r k e r s A s part of the mounting process, six markers were placed along the circumference of the mitral annulus. See Figure 2.8 for a depiction o f the marker locations. Also , the markers are visible in the picture of the annulus seen in Figure 2.9. Five of the six markers consisted o f small stainless steel pins shortened to a length of approximately 4 mm. The pins were set into the annulus perpendicular to the plane o f the valve orifice. The five pins were placed in specific locations around the annulus. One pin is located at each trigone, and the other three are spread across the posterior section of the annulus. One pin was placed in the annulus directly in the middle of P2. The other two pins were placed at the approximate middles of PI and P3. The remaining marker was placed equidistant from the two trigones; on the anterior annulus at the middle of the anterior leaflet. It was not possible to use a pin for the marker at the middle of the anterior leaflet because there is not enough tissue to hold the pin. The pin would interfere with the aorta. Instead a small suture knot was used as a marker. The purpose for using this specific orientation of markers is to allow documentation and calculation of movement along specific points of the annulus. Also , it is possible to calculate the circumference and area of the annulus using data collected 29 from the six markers. Furthermore, the anterior-posterior (A-P) distance of the valve can be determined by measuring the distance between the markers 2 and 5. This is important for the experiment because the A - P distance is used to determine the magnitude of annular constriction. Figure 2.8 - The location and number of markers placed along the annulus. 30 Figure 2.9 - A picture of the exposed mitral valve and annulus with markers in place. Also pictured are the device modules implanted on the outside surface of the heart. 2.7 Device and Module design 2.7.1 Design Cr i t e r i a In order to test the hypothesis that surgical manipulation o f the annulus alone is capable of reducing mitral regurgitation, it was necessary to design and build a device to implant on the heart near the annulus. The purpose for this device is to change the shape and size of the annulus as required for the experiments. A s part of the design, the following attributes were required: 1) the device must apply even and consistent force to the posterior section o f the annulus; 2) the device must not interfere with any major coronary vessels; 3) the device must be relatively quick and easy to implant; 4) the device must be capable of implantation on any heart. The device must apply even force across the posterior annulus because only when doing so is it possible to produce an even and symmetrical constriction across the 31 annulus. If the force is not distributed evenly, annular constriction w i l l be asymmetric and leaflet coaptation could be affected. In order to maintain clinical relevancy, the device must avoid interference with major coronary vessels. For this study arteries and veins with diameters equal to or greater than 1.0 mm were considered significant and therefore could not be obstructed or ligated by the device. It was advantageous to design a device which could be implanted quickly and easily because a simple implantation procedure enables the device to be consistently implanted for all trials in the study. N o two hearts are exactly the same. Among other traits, there are always minor anatomical differences in heart size, annulus size, anterior and posterior leaflet orientation, and coronary vessel arrangement. The orientation and size of the circumflex artery and the distal branches can differ significantly between hearts. The device must be able to accommodate these anatomical differences without compromising any of the other required traits. 2.7.2 Device Design Overview The device used in this study consists of four modules implanted onto the exterior surface of the heart. The modules are implanted along the posterior section of the mitral annulus. The modules are linked together through two transverse sutures running parallel with the plane of the annulus. Figures 2.9 and 2.10 are pictures of the device implanted on a porcine heart. The two transverse sutures apply tension to the modules, and thus are referred to as the tensioning sutures. The tensioning sutures are anchored into the left ventricle muscle wall . The ventral anchor is placed near the left descending coronary artery ( L D A ) , and the dorsal anchor is placed near the right coronary artery ( R C A ) . The protocol for device implantation w i l l be described in section 2.7.6. The function of the device is a result of the tensioning sutures causing a purse-string effect on the four modules. When the tensioning sutures are shortened, the length of the posterior section of the mitral annulus is reduced. The posterior section of the annulus also moves inward towards the anterior annulus, reducing the anterior-posterior (A-P) distance o f the mitral valve. 32 Figure 2.10 - The device modules implanted on the outside surface of the heart. 2.7.3 Design of the Standard Modules The basic shape of a standard module can be seen in Figure 2.11. A standard module is 4 mm wide, 10 mm long, and has a height of 6 mm. In this study, all modules had a beveled posterior section. The bevel was incorporated into the module design with the intention of reducing the footprint area of the module while allowing the lower tensioning suture to work as a cantilever for the purpose o f modulating the attitude of the modules. The upper tensioning suture travels though the module at the middle of the footprint area. The lower tensioning suture travels through the module near the tip of the module over the bevel. With this arrangement, it is possible to control the attitude of the modules by increasing or reducing the tension in the lower tensioning suture. Increased tension in the lower tensioning suture causes the modules to pitch upward; reduced tension causes the modules to pitch downward. Figure 2.12 depicts the attitude modulation o f the modules. 33 Side View o o Standard Module Bottom View O o O o Top View Figure 2.11 - The side, bottom, and top views of the standard module design. Attitude Adjustment of the Modules Balanced tensioning sutures Bottom tensioning suture Top tensioning suture applying too applying too much force much force Figure 2.12 - Attitude adjustment of the modules is performed though tensioning of the top and bottom transverse sutures. 2.7.4 Modules wi th Alternative Footprints In order to accommodate implantation on every heart, it was necessary to design a variety of modules with different shapes. This was necessary because variation in heart anatomy, specifically the orientation of the coronary vessels, made it impossible to 34 exclusively implant standard modules without vascular interference. A l l modules retained similar basic morphology with identical lengths, heights, and tensioning suture locations. The feature of variation in the modules was the width and the specific form of the module footprint. When a standard module would be in conflict with a coronary vessel, an alternative module was selected which could bridge over the conflicting vessel. These alternative modules have channels removed from the footprint in which the vessels can lie undisturbed. Figure 2.13 depicts the variety of module footprints used in this study. It was not required to custom make modules to fit a specific heart. Instead, pre-made modules would be selected based on their ability to fit a particular heart. Multiple copies of each footprint were available for each experiment. Standard Module t — >s t 1 1 f : - N Figure 2.13 - Footprints of the standard and alternative modules. 2.7.5 Construction of the Device The modules were individually handmade. A l l modules were machined from a single piece of polyethylene plastic. A high speed steel cutting bit was used in a high speed rotary tool to carve the plastic. Polyethylene was selected because of its high 35 strength and ease in which it could be carved. This material is rigid enough to be strong, but pliable enough to resist cracking when machined. 2.7.6 Implantation of the Modules Figure 2.11 also depicts the location of the various suture holes drilled into the standard modules. A s discussed previously the tensioning suture holes travel transversely through the module. Four holes placed along the corners of the footprint are used to suture the module to the external surface of the heart. A t each of the corner holes, a 2-0 silk suture was passed through the hole and superficially into the tissue immediately below. This provided adequate strength for module implantation. Since the force applied by the tensioned modules w i l l be straight down into the tissue, it is not necessary to hold the modules in place with a large number of sutures. The downward force of the modules helps keep them in place. Also , there are no lateral pulling forces to tear the superficial sutures from the tissue. 2.7.7 Module Location across the Annulus The location for implantation of the modules is of utmost importance. Failure to implant the modules correctly could lead to asymmetrical constriction of the annulus. The ideal location for the modules is on the exterior surface of the heart directly over the annulus. For the porcine heart this location is slightly apical to the circumflex artery where the cleft begins between the ventricle and atrial appendage. A l l modules must be implanted in the same plane as the annulus and be evenly spaced to achieve an even distribution of constricting force. If the ideal location of a standard module conflicted with a coronary vessel equal to or greater than 1.0 mm in diameter, then an alternative module capable o f bridging the vessel was selected. In all trials, it was possible to find appropriate alternative modules to fit the ideal locations. If a standard module conflicted with a coronary vessel with a diameter less than 1.0 mm, the module was placed directly over the vessel. 2.7.8 Tensioning Suture Implantation The tensioning sutures consisted of 2-0 silk anchored into the ventricle wall as close to the left descending artery and right coronary artery as possible. The anchors are required to hold the suture tensioned without tearing the tissue. Three deep throws into 36 the muscle wall produced sufficient holding strength. A s with the modules, the anchors were placed in the plane of the annulus. 2.8 The Model of Mitral Valve Regurgitation 2.8.1 Creation of Type II and Type Ill-b Mitral Valve Dysfunction A novel in-vitro model of mitral valve disease was used during this study to simulate three lesions producing two types of dysfunction. Regurgitation was created through the creation of a type Il l-b restricted posterior leaflet due to shortened chords, a type II prolapsed posterior leaflet due to lengthened chords, and a type II flail posterior leaflet due to cut chords. Marginal chords are defined as those inserting directly into the free edge o f the leaflet [79]. During this study, only marginal chords connected to P2 of the posterior leaflet were surgically altered. 2.8.2 Creation of the Restricted Posterior Leaflet The restricted leaflet was formed through shortening of chords connected to the posterior leaflet. The papillary muscles were not altered, neither was the size or orientation of the annulus. The posterior leaflet was restricted in motion due to the increased tethering force applied by the chords. Wi th a porcine heart mounted in the heart cage, the left ventricle lumen was evacuated of any water. The largest marginal chord inserting into P2 from the anterolateral papillary muscle and the largest marginal chord inserting into P2 from the posteromedial papillary muscle were selected for alteration. The lengths o f these two chords were measured using digital calipers. The length was measured from the insertion point at the posterior leaflet to the insertion point on the papillary muscle. The chords were then cut at the approximate middle. To simulate shortened chords, artificial chords shorter than the natural chords were created. Figure 2.14 is an illustration of the artificial chords tethering the posterior leaflet. The artificial chords used the same insertion points as the natural chords. In order to create artificial chords, it was beneficial to make an anchor on the tip of the papillary muscle through which the artificial chord would be attached. To make the anchor, a small loop approximately 2 mm in diameter was tied in 2-0 silk suture. The 37 loop was sutured to the papillary muscle in the exact location of the chord insertion point. The fibrous insertion point on the papillary muscle was strong enough to hold the anchor loop in place; even when tensioned during ventricular pressurization. Throughout the study, there were no cases of anchors tearing loose. The artificial chord was made using 2-0 silk sutures. The suture was passed through the top of the leaflet at the natural chordal insertion point. The suture then traveled down through the anchor loop and back up through the leaflet in the same approximate location of the chordal insertion point. The two insertion points of the artificial chord were approximately 2-3 mm apart from each other. The length of the artificial chord was adjusted by clamping a hemostat under the leaflet at the chordal length desired. The intended length of the shortened chord was approximately 5 mm shorter than the natural chord. The two ends of the artificial chord were then tied at that length. The length of the artificial chord was then measured. This entire process would then be repeated for the other chord selected to be altered. 38 Annulus Figure 2.14 - Artificial chords were created using 2-0 sutures anchored to the papillary muscle. 2.8.3 Creation of a Prolapsed Leaflet To create the prolapsed leaflet, two marginal chords inserting into the P2 portion of the posterior leaflet were replaced with artificial chords of increased length. In order to create the artificial elongated chords, the same methodology as described for the shortened chords was used. During this study the chords were elongated to a length which allowed the posterior leaflet to prolapse approximately 5 mm above the plane of the anterior leaflet. In all cases, the lengths of the chords were measured before and after manipulation. 39 2.8.4 Creation of Ruptured Chords To simulate ruptured chords, the chords inserting into the P2 portion of the posterior leaflet were cut. The number of chords cut was determined by the ability to produce a flail leaflet. For the majority of the trials, the cutting of one medial marginal chord and one lateral marginal chord was sufficient to produce a flail P2. However, there were instances were more chords were required to be cut. Before cutting, all chord lengths were measured with digital calipers for the purpose of comparison with the shortened and elongated chord lengths. 2.9 Experimental Protocol 2.9.1 Overview During this study, ten porcine hearts were used to perform the preceding experiments. A t the beginning of each experiment, the heart was mounted in the heart cage and the device was implanted as described previously. Using the heart cage, the left ventricle was pressurized to 107 mmHg. Baseline measurements of the regurgitation flow rate were performed on the untouched mitral valve apparatus. The regurgitant volume was collected during a period of five minutes. The volume was measured to the nearest 25 milliliters. This was used to calculate the regurgitation flow rate in milliliters per minute. Each heart was used to create a flail leaflet, prolapsed leaflet, and a restricted leaflet. For each created lesion, four measurements of regurgitation flow rate were performed. In sequence, the regurgitation flow rate of the created disease, 10% annular constriction, 30% annular constriction, and 50% annular constriction were measured. In total, thirty mitral valve lesions were created and ninety trials of annular constriction were performed. 2.9.2 Measurement of Annulus Markers During every five minute period when the regurgitant volume was being collected, the distances between the six markers placed along the circumference of the annulus were measured. Measurements were performed with digital calipers and are accurate to +/- 0.5 mm. The distances measured were between markers: 1 and 2, 2 and 3, 40 3 and 4, 4 and 5, 5 and 6, 6 and 1, 2 and 6, 2 and 5, 2 and 4. Refer to Figure 2.8 for the location and number of each marker along the annulus. The most critical measurement of baseline annular dimensions is the distance between markers 2 and 5. This is the A - P distance of the mitral valve. The A - P distance was used to measure the magnitude of annular constriction. If the A - P distance was reduced by half, then the annulus was said to be 50% constricted. The baseline A - P distance was used to calculate A - P distances for annular constrictions of 10%, 30%, and 50%). The calculated A - P distances were used as "target" distances for the 2 and 5 markers. During experimentation, the annulus would be constricted until the 2 and 5 markers reached the calculated target distance for the desired constriction. In all cases, it was possible to manipulate the annulus to create a constriction resulting in an A - P distance within one millimeter of the calculated target. 2.9.3 Calculation of the Annulus Area The annulus area was calculated using the general formula for the area of an ellipse: area = V2 long axis * Vi short axis * P i . Since the annulus closely resembles a D shape, it is essentially one half of an ellipse. Therefore, the area of the annulus was calculated as a full ellipse, and then divided by two to find the actual area of the D shaped annulus. Although it does not accommodate the exact geometric conformation of the annulus, it is believed that this calculation represents the area o f the annulus well enough for this study. For the actual calculation, the long axis was calculated by adding the distances from markers 1 to 2, and 2 to 3. This is essentially the distance across the anterior annulus from trigone to trigone. The complete long axis was then halved. The V2 short axis was the distance between markers 2 and 5; this is the A - P distance. The annulus area was calculated using the following formula: [((1 to 2) + (2 to 3))/2*(2 to 5)*Pi]/2 2.9.4 Measurement of the Disease Regurgitation Flow Rate After creation of each valvular lesion, the left ventricle was pressurized and the regurgitant volume was captured and measured, enabling the calculation of the disease regurgitation flow rate. The normal duration for ventricular pressurization and regurgitation was five minutes. On occasion when regurgitation was extraordinarily 41 severe the duration would be shorter because the receptacle for the regurgitant volume would f i l l to capacity in less than five minutes. For this reason several disease regurgitant rates were calculated using experimental durations of one or two minutes. 2.9.5 Qualitative Assessment of Initial Disease Regurgitation Flow Rate In conjunction with the quantitative measurements, a qualitative assessment of the initial disease regurgitation rate was performed for each lesion. The regurgitant flow rate was deemed mild, moderate, or severe based on qualitative observations of the regurgitant jet magnitude and regurgitation flow rate. Qualitatively assessed mild, moderate, and severe regurgitation roughly correlate with quantitatively assessed regurgitation flow rates of less than 500 ml/min, between 500 and 1500 ml/min, and over 1500 ml/min, respectively. 2.10 Statistical Analysis Statistical analysis was performed primarily using J M P i n software developed by S A S Institute Inc. Games-Howell tests were performed using SPSS 11.0 software. In all cases, p values less than 0.05 were considered significant. In order to correctly perform an analysis of variance ( A N O V A ) statistical test, there must be equal variance between groups [91]. Homogeneity of variance is tested using a Levene test [92]. A l l data groups were subjected to a Levene test before performing an A N O V A . If the data had equal variance, then a standard A N O V A was applied. In the majority of cases the data did not have equal variance between groups. A Welch A N O V A is similar to a standard A N O V A except it does not make an assumption of homogeneous variance and therefore can be used for analysis of data with unequal variance [93]. Welch A N O V A tests were performed on all data groups with unequal variance. If a Welch A N O V A test found a significant difference between groups, then a Games-Howell test was performed to determine which groups were different from each other. The Games-Howell test is a multiple comparison test which can be used for analysis o f data with unequal variance [94]. 42 Chapter 3 - Results 3.1 Analysis of the Mitral Regurgitation Model 3.1.1 Changes in the Chord Length The average change in length for shortened and elongated chords is displayed in Table 3.1. For the creation of a type Il l-b restricted leaflet, chords were shortened an average of 6.7 mm. To create a type II prolapsed leaflet, chords were elongated an average of 6.0 mm. Alteration Performed Number of Chords Altered Average Change in Chord Length (mm) Standard Deviation (mm) Chord Shortened 20 -6.7 1.7 Chord Lengthened 20 6.0 2.0 Table 3.1 - The average change in chord length after chord shortening and chord lengthening procedures. 3.1.2 Changes in the Anterior-Posterior (A-P) Distance The A - P distance is the distance from the anterior annulus at A 2 to the posterior annulus at P2. Using the markers placed along the annulus, the A - P distance is the measured distance between markers 2 and 5. Figure 3.1 depicts the measured A - P distance for each level of annular constriction. The average A - P distance for the baseline heart, 10% annular constriction, 30% annular constriction, and 50% annular constriction was 22.5, 20.1, 15.9 and 11.5 mm, respectively. 43 25 20 E E 15 o o c JS (A Q Q. 10 I < 0 . . . . . Baseline 10% 30% Annulus Constriction 50% Figure 3.1 - Average measured distance across the mitral valve (A-P distance) (mm) with standard error for the baseline and constricted annulus. 3.1.3 Changes in the Annulus A r e a Figure 3.2 depicts the average annulus area for the baseline heart and all levels of annular constriction. The average annulus area for the baseline heart, 10% annular constriction, 30% annular constriction, and 50% annular constriction was 678.2, 600.5, 466.4 and 332.1 mm 2 , respectively. 44 800 700 600 <•£ 500 E < 400 (A C < 300 200 100 B a s e l i n e 10% 30% A n n u l u s Cons t r i c t ion 50% Figure 3.2 - Average calculated annulus area (mm ) with standard error for the baseline and constricted annulus. 3.1.4 Changes in the Annulus Circumference The annulus circumference is the sum of the measured distances between the annulus markers placed along the circumference o f the annulus. The average annulus circumference for the baseline and all levels of annular constriction are depicted in Figure 3.3. The average annulus circumference for the baseline heart, 10% annular constriction, 30% annular constriction, and 50% annular constriction was 108.7, 102.9, 95.2 and 87.5 mm, respectively. 45 120 100 E E o o c 2? £ E o It) c < 80 60 40 20 Baseline 10% 30% A n n u l u s Const r ic t ion 50% Figure 3.3 - Average measured annulus circumference (mm) with standard error for the baseline and constricted annulus. 3.1.5 M e a s u r e d C h a n g e s i n the D i s t a n c e s between A n n u l u s M a r k e r s Measurement o f the distance between annulus markers provides information about the changes to specific regions of the annulus. The average change in distance between markers for each segment o f the annulus is represented in Figure 3.4. Refer to Figure 2.8, for the marker locations with respect to the annulus. Markers 1 to 2 and 2 to 3 cover the anterior annulus essentially from trigone to trigone. When the annulus was constricted, the reduction in distance between markers 1 and 3 was relatively minimal. In all cases, the reduction in distance between 1 and 2 or 2 and 3 was less than 2.0 mm. 46 Across the rest of the circumference o f the annulus, from marker 3 to 4, 4 to 5, 5 to 6, and 6 to 1, the reduction in distance between the markers was substantially greater. The greatest reduction in distance between markers occurred between 2 to 4, 2 to 5, and 2 to 6. These are the measurements which spanned across the valvular orifice; with the 2 to 5 measurement being the A - P distance. 11.0 _ 9.0 E E 7.0 52 ro c 0) Q) 0 CO. 0 o c a in Q c a> o> 3.0 ro .c O 5.0 1.0 -1.0 I I 1 __> 2 2 -> 3 3 -> 4 4 --> 5 5 -> 6 6 -> 1 2 -> 6 2 -> 5 2 -> 4 H 1 0 % Constriction • 30% Constriction • 50% constriction Segment Along the Annulus Figure 3.4 - Average change in distance between markers along the annulus for annular constrictions of 10%, 30%, and 50%. 47 3.1.6 Regurgitation Flow Rate for the Baseline and created Lesions The regurgitation flow rates for the normal baseline, flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords are presented in Figure 3.5. The created lesions produce regurgitation flow rates greater than the baseline value. A Welch A N O V A test produces p < 0.0001 indicating that there is a significant difference between the means. 48 Disease Regurgitation Welch ANOVA p < 0.0001 2500.0 2000.0 c £ o LL C o 3 E? 3 O) <D 01 1500.0 1000.0 500.0 0.0 2107.1 2221.6 588.9 I Baseline (No Disease) Cut Chords Lengthened Chords Shortened Chords Disease Type Figure 3.5 - The regurgitation flow rate (ml/min) for the baseline heart and the flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords. A Welch ANOVA test produces p O .0001 . Error bars indicate standard error. 49 3.1.7 Games-Howell test of the Regurgitation Flow Rate for the Baseline and Created Lesions A Games-Howell test was performed as a post-hoc multiple comparison of the means of the regurgitation flow rate for normal baseline, flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords The results of the test are presented in Table 3.2. The means are significantly different for the baseline and cut chords groups (p = 0.030), the baseline and lengthened chords groups (p = 0.007) and the baseline and shortened chords groups (p = 0.012). Also , the means are significantly different for the lengthened chords and shortened chords groups (p = 0.041). N o other comparison between groups shows significantly different means. Comparison Significant Difference Between Means p Value Baseline Cut Chords Yes 0.030 Lengthened Chords Yes 0.007 Shortened Chords Yes 0.012 Cut Chords Baseline Yes 0.030 Lengthened Chords N o 0.999 Shortened Chords N o 0.131 Lengthened Chords Baseline Yes 0.007 Cut Chords N o 0.999 Shortened Chords Yes 0.041 Shortened Chords Baseline Yes 0.012 Cut Chords N o 0.131 Lengthened Chords Yes 0.041 Table 3.2 - Games-Howell multiple comparison test of regurgitation flow rates for the normal baseline, flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords. 3.2 The Effect of Annular Constriction on Mitral Regurgitation 3.2.1 The Effect of Annular Constriction on a Flail Leaflet due to Cut Chords The average regurgitation flow rate (ml/min) for the flail leaflet and the annular constrictions of 10%, 30% and 50% are depicted in Figure 3.6. Annular constriction produces a reduction in the regurgitation flow rate. A Welch A N O V A test produces p = 0.0054 indicating that there is a significant difference between the means. 50 Cut Chords Welch ANOVA p = 0.0054 2500.0 2000.0 c | i 1 1500.0 u_ c o s ra u 3 o? 1000.0 or 500.0 2107.0 731.0 253.6 0% (Baseline) 0% (Disease) 10% 30% Annular Constriction .34.1 50% Figure 3.6 - Regurgitation flow rates (ml/min) for the baseline and cut chords with annular constriction o f 0% (disease), 10%, 30%, and 50%. Welch A N O V A test of 0% (disease), 10%, 30%, and 50% annular constriction produces p = 0.0054. Error bars indicate standard error. 51 3.2.2 Games-Howell test of the Effect of Annular Constriction on a Flail Leaflet A Games-Howell test was performed as a post-hoc multiple comparison o f the means of the regurgitation flow rate for the flail leaflet and annular constrictions of 10%, 30%, and 50%. The results of the test are presented in Table 3.3. The means are significantly different for the 0% (Disease) and 50% annular constriction groups (p = 0.032). N o other comparison between groups shows significantly different means. Comparison Significant Difference Between Means p Value 0% (Disease) 10%o Constriction N o 0.216 30% Constriction N o 0.056 50% Constriction Yes 0.032 10% Constriction 0% (Disease) N o 0.216 30% Constriction N o 0.432 50%o Constriction N o 0.124 30% Constriction 0% (Disease) N o 0.056 10% Constriction N o 0.432 50% Constriction N o 0.395 50% Constriction 0% (Disease) Yes 0.032 10% Constriction N o 0.124 30% Constriction N o 0.395 Table 3.3 - Games-Howell multiple comparison test of regurgitation flow rates for 0% (disease), 10%, 30%>, and 50% annular constriction for hearts with cut chords. 3.2.3 The Effect of Annular Constriction on a Prolapsed Leaflet due to Lengthened Chords The average regurgitation flow rate (ml/min) for the prolapsed leaflet and the annular constrictions of 10%, 30% and 50% are depicted in Figure 3.7. Annular constriction produces a reduction in the regurgitation flow rate. A Welch A N O V A test produces p = 0.0018 indicating that there is a significant difference between the means. 52 Lengthened Chords Welch A N O V A p = 0.0018 2500.0 2000.0 c I i 1500.0 Li. c o a E? ^ 1000.0 500.0 0.0 0% (Baseline) 0% (Disease) 10% Annular Constriction 30% 50% Figure 3.7 - Regurgitation flow rates (ml/min) for the baseline and lengthened chords with annular constriction of 0% (disease), 10%, 30%, and 50%. Welch A N O V A test of 0%> (disease), 10%, 30%, and 50% annular constriction produces p = 0.0018. Error bars indicate standard error. 53 3.2.4 Games-Howell test of the Effect of Annular Constriction on a Prolapsed Leaflet A Games-Howell test was performed as a post-hoc multiple comparison of the means of the regurgitation flow rate for the prolapsed leaflet and annular constrictions of 10%, 30%), and 50%. The results o f the test are presented in Table 3.4. The means are significantly different for the 0% (Disease) and 30% annular constriction groups (p = 0.016). Also , the means are significantly different for the 0% (Disease) and 50% annular constriction groups (p = 0.008). N o other comparison between groups shows significantly different means. Comparison Significant Difference Between Means p Value 0% (Disease) 10%> Constriction N o 0.264 30%o Constriction Yes 0.016 50% Constriction Yes 0.008 10% Constriction 0% (Disease) N o 0.264 30% Constriction N o 0.418 50% Constriction N o 0.181 30%o Constriction 0% (Disease) Yes 0.016 10% Constriction N o 0.418 50% Constriction N o 0.410 50% Constriction 0% (Disease) Yes 0.008 10% Constriction N o 0.181 30% Constriction N o 0.410 Table 3.4 - Games-Howell multiple comparison test of regurgitation flow rates for 0%> (disease), 10%, 30%, and 50% annular constriction for hearts with lengthened chords. 54 3.2.5 The Effect of Annular Constriction on a Restricted Leaflet due to shortened Chords The average regurgitation flow rate (ml/min) for the restricted leaflet and the annular constrictions of 10%, 30% and 50% are depicted in Figure 3.8. Annular constriction produces a reduction in the regurgitation flow rate. A Welch A N O V A test produces p = 0.0011 indicating that there is a significant difference between the means. 55 Shortened Chords Welch ANOVA p = 0.0011 700.0 600.0 ^ 500.0 c | E, o 400.0 Li. C o ro +-* CD or 300.0 200.0 100.0 0.0 588.9 16.5 281.0 .60.5 36.0 0% (Baseline) 0% (Disease) 10% 30% An n u la r C o nstrictio n 50% Figure 3.8 - Regurgitation flow rates (ml/min) for the baseline and shortened chords with annular constriction of 0% (disease), 10%, 30%, and 50%. Welch A N O V A test of 0% (disease), 10%, 30%, and 50% annular constriction produces p = 0.0011. Error bars indicate standard error. 56 3.2.6 Games-Howell test of the Effect of Annular Constriction on a Restricted Leaflet A Games-Howell test was performed as a post-hoc multiple comparison of the means of the regurgitation flow rate for the restricted leaflet and annular constrictions of 1 0 % , 30%o, and 5 0 % . The results of the test are presented in Table 3 . 5 . The means are significantly different for the 0 % (Disease) and 3 0 % annular constriction groups (p = 0 . 0 1 9 ) . Also , the means are significantly different for the 0 % (Disease) and 5 0 % annular constriction groups (p = 0 . 0 1 5 ) . N o other comparison between groups shows significantly different means. Comparison Significant Difference Between Means p Value 0 % (Disease) 1 0 % Constriction N o 0 . 2 5 2 3 0 % Constriction Yes 0 . 0 1 9 50%) Constriction Yes 0 . 0 1 5 10%> Constriction 0 % (Disease) N o 0 . 2 5 2 30%) Constriction N o 0 . 0 5 2 5 0 % Constriction Yes 0 .031 3 0 % Constriction 0%> (Disease) Yes 0 . 0 1 9 1 0 % Constriction N o 0 . 0 5 2 50%o Constriction ' N o 0 . 5 5 7 5 0 % Constriction 0%> (Disease) Yes 0 . 0 1 5 10%) Constriction Yes 0 .031 30%) Constriction N o 0 . 5 5 7 Table 3 .5 - Games-Howell multiple comparison test of regurgitation flow rates for 0 % (disease), 1 0 % , 3 0 % , and 5 0 % annular constriction for hearts with restricted leaflets due to shortened chords. 5 7 3.3 The Effect of the Lesion Type To evaluate whether the specific type of leaflet pathology responds differently to annular constriction, we compared the M R reduction percentage at each magnitude o f constriction, for each type of disease. 3.3.1 The Effect of Lesion Type on the Ability to Reduce Mitral Regurgitation at 10% Annular Constriction The regurgitation reduction (%) at 10% annular constriction is compared between the flail leaflet, prolapsed leaflet, and restricted leaflet in Figure 3.9. Analysis of variance testing produces p = 0.1537. There is no significant difference in means for any lesion type at 10%> annular constriction. 58 100.0 90.0 10% Annular Constriction ANOVA p = 0.1537 80.0 5 70.0 o Flail (cut) Prolapsed (lengthened) Restricted (shortened) Disease Type Figure 3.9 - Regurgitation reduction (%) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords at 10% annular constriction. Analysis of variance test produces p = 0.1537. Error bars indicate standard error. 3.3.2 The Effect of Lesion Type on the Ability to Reduce Mitral Regurgitation at 30% Annular Constriction The regurgitation reduction (%) at 30% annular constriction is compared between the flail leaflet, prolapsed leaflet, and restricted leaflet in Figure 3.10. Analysis of 59 variance testing produces p = 0.5527. There is no significant difference in means for any lesion type at 30% annular constriction. 30% Annular Constriction ANOVA p = 0.5527 100.0 T -90.0 80.0 I 70.0 O Flail (cut) Prolapsed Restricted (shortened) (lengthened) Disease Type Figure 3.10 - Regurgitation reduction (%) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords at 30% annular constriction. Analysis of variance test produces p = 0.5527. Error bars indicate standard error. 60 3.3.3 The Effect of Lesion Type on the Ability to Reduce Mitral Regurgitation at 50% Annular Constriction The regurgitation reduction (%) at 50% annular constriction is compared between the flail leaflet, prolapsed leaflet, and restricted leaflet in Figure 3.11. Welch A N O V A testing produces p = 0.2427. There is no significant difference in means for any lesion type at 50% annular constriction. 61 c o 100.0 90.0 80.0 70.0 50% Annular Constriction Welch ANOVA p = 0.2427 CD DC c o 2 e 3 O d) 60.0 50.0 40.0 30.0 20.0 Flail (cut) Prolapsed (lengthened) Restricted (shortened) Disease Type Figure 3.11 - Regurgitation reduction (%) for flail leaflet due to cut chords, prolapsed leaflet due to lengthened chords, and restricted leaflet due to shortened chords at 50% annular constriction. Welch A N O V A test produces p = 0.2427. Error bars indicate standard error. 3.4 The Effect of the Init ial Disease Severity In order to test the effect of the initial disease severity on the ability of annular constriction to reduce M R , we compared the regurgitation flow rate for mild, moderate, and severe disease at annular constrictions of 0%> (disease), 10%>, 30%, and 50%. 62 3.4.1 The Effect of the Initial Disease Severity on the ability to Reduce Regurgitation through Annular Constriction In Figure 3.12, the average regurgitation flow rate for the disease, 10% annular constriction, 30% annular constriction, and 50% annular constriction are presented with the initial disease severity. A s would be expected, for mild , moderate, and severe disease severities, increased annular constriction produces a reduction in the regurgitation flow rate. A t mild and moderate initial disease severities, annular constriction of 30% or 50%) reduces the regurgitation flow rate to very low, essentially baseline levels. When the initial disease is rated severe, 30% annular constriction reduces M R a great deal but does not reduce M R to baseline. Only 50% annular constriction reduces severe disease to baseline M R values. 63 3000 2500 c J i 2000 o LL o 2 5) k_ 3 O) Q) or 1500 1000 500 m 0% (Disease) @l 10% Constriction • 30% Constriction • 50% Constriction Mild Moderate Severe Severity of Disease Before Annuloplasty Figure 3.12 - Average measured regurgitation flow (ml/min) for hearts with mild, moderate, or severe mitral regurgitation before annular constriction. Annular constrictions of 0% (Disease), 10%, 30%>, and 50%o are represented. Error bars indicate standard error. 3.4.2 Welch A N O V A Testing of the Effect of Disease Severity Welch A N O V A tests were performed to determine i f there are significant differences in regurgitation flow rates between groups with equal annular constriction but 64 differing initial disease severities. Initial disease severity was found to have a significant effect on the regurgitation flow rates at 10% and 30% annular constriction. A t 50% annular constriction there was not a significant difference in regurgitation flow rate regardless of the initial disease severity. Magnitude of Annular Constriction Initial Disease Severity has a Significant Effect on the Abi l i ty to Reduce M R through Annular Constriction Welch A N O V A p Value 10% Yes 0.0064 30% Yes 0.0270 50% N o 0.3585 Table 3.6 - Welch A N O V A tests of the effect of initial disease severity on the ability to reduce M R through annular constriction of 10%, 30%, and 50%. 3.5 Qualitative Observations of the Mitral Valve during Annular Constriction A t 10% annular constriction the regurgitant jet typically remains present but is greatly reduced in size. The posterior leaflet spans the valve orifice several millimeters, similar to a normal valve. A t 30%> annular constriction the regurgitant jet is typically reduced to a small upwelling or ripple. The posterior leaflet is reduced in span across the annulus as the coaptation point with the anterior leaflet is only a couple millimeters from the posterior annulus. The posterior leaflet would still partially functional at this magnitude of constriction. A t 50% annular constriction the regurgitant jet is typically eliminated. The valve is essentially operating as a monoleaflet. The anterior leaflet essentially coapts with the posterior annulus. The posterior leaflet does not span the orifice significantly and is essentially nonfunctional. 65 Chapter 4 - Discussion 4.1 Overview The in-vitro model of mitral regurgitation used in this study produced regurgitation through the creation of three distinct lesions. Cut chords created a flail leaflet, lengthened chords created a prolapsed leaflet, and shortened chords created a restricted leaflet. The device developed for this study effectively constricted the annulus in a precise and graded manner. Annular constriction was found to reduce M R for the flail leaflet, prolapsed leaflet, and restricted leaflet. In all cases, annular constriction alone without any other surgical intervention was capable o f reducing M R significantly. The type of lesion did not have a significant effect on the amount of regurgitation reduction for any given annular constriction. The magnitude of annular constriction was the only variable found to affect the efficiency of M R reduction by annuloplasty. The severity o f regurgitation before annular constriction affects the regurgitation flow rate after annular constriction only when the magnitude of constriction is 10% or 30%. When the annulus was constricted 50%, the regurgitation flow rate does not differ between groups with different initial disease severity. 4.2 The Reduction of Mitral Regurgitation through Annular Constriction 4.2.1 Hypothesis of Annular Constriction The main purpose of this study is to investigate whether constriction of the annulus alone is capable of reducing M R . The hypothesis tested is: "For type II and type Il l-b mitral dysfunction associated with the posterior leaflet, constriction of the annulus with no other surgical alteration to the mitral valve apparatus w i l l be sufficient in producing a reduction in mitral valve regurgitation due to an improvement in leaflet coaptation and formation of a monoleaflet valve". The data produced by measuring the regurgitation flow rate for the flail leaflet, prolapsed leaflet, and restricted leaflet is used to test this hypothesis. 4.2.2 Analysis of the Regurgitation Flow Rates The regurgitation flow rates for the created lesions and the subsequent annular constrictions of 10%, 30%, and 50% are presented in Figures 3.6, 3.7, and 3.8 for the flail 66 leaflet, prolapsed leaflet, and restricted leaflet, respectively. For each lesion a Welch A N O V A test indicates a significant difference in means between groups. Annular constriction produced a significant decrease in the regurgitation flow rate. The post-hoc Games-Howell test presented in Tables 3.3 confirms that for the flail leaflet there is a significant difference between the regurgitation rate of the initial disease and 50% annular constriction. Similarly, as seen in Table 3.4, the prolapsed leaflet regurgitation flow rate at 30%> and 50%> annular constriction is significantly different from the initial disease flow rate. Concluded from the data in Table 3.5, the restricted leaflet also has significant differences between the initial disease regurgitation flow rate and the regurgitation flow rate at both 30% and 50% annular constriction. The Welch A N O V A and Games-Howell tests confirm that for a flail leaflet due to cut chords, a prolapsed leaflet due to lengthened chords, and a restricted leaflet due to shortened chords annular constriction alone is capable of reducing M R . Without any other surgical alteration to the mitral valve apparatus, the regurgitation flow rate decreased for each increase in annular constriction. For a l l three lesions, 50% annular constriction is associated with very low regurgitation flow rates. The regurgitation flow rates at 50% constriction are comparable to the regurgitation flow rates at baseline before creation of a lesion. Regurgitation due to the lesion was virtually eliminated. 4.3 The Effect of Annular Constriction on Type Ill-b Dysfunction 4.3.1 Evidence Supporting the Importance of Annular Constriction for Type Ill-b Dysfunction The ability to reduce M R due to type Il l-b dysfunction through annular constriction alone has also been determined experimentally through a series of experiments performed at the Stanford University School of Medicine [71-73, 95, 96]. Timek et al use the underlying principle of "septal-lateral cinching" to constrict the annulus in an effort to reduce M R [73]. For the sake of clarity, the septal-lateral cinching referred to by Timek et al is simply a reduction in the A - P distance. The terminology differs between the clinical references used throughout this study and the ovine model references used by Timek et al. The studies of septal-lateral annulus cinching were performed on live ovine subjects under full cardio-pulmonary bypass [95]. Temporary 67 ligation of the proximal left circumflex artery was used to create acute ischemic M R with type Il l-b dysfunction [73]. In another study, permanent ligation o f the second and third obtuse marginal coronary arteries was used to create chronic M R with type Il l-b dysfunction [71]. Regurgitation was measured on the beating heart using echocardiography [73]. Constriction of the annulus was achieved through the placement of a single Prolene suture running from an anchor in the anterior annulus through the posterior annulus and externalized to the epicardial surface [73]. A tourniquet was used to tension the suture thereby cinching the annulus and decreasing the A - P distance [73]. There was no direct surgical manipulation of the leaflets or any other components of the mitral valve apparatus. It was found that septal-lateral cinching of the annulus was capable of decreasing M R to the measured baseline values [72, 73, 96]. Although the models are different, these series of experiments produced results comparable to those found in this study; annular constriction without any other surgical intervention significantly reduced M R for the restricted leaflets with type Il l-b dysfunction. To our knowledge, the method of septal-lateral cinching, as described by Timek, has not been applied to type II dysfunction such as a flail or prolapsed leaflet. 4.3.2 Annular Constriction Using the Coapsys Device Based on the same general principle that annular constriction alone can reduce M R for type Il l-b dysfunction, Myocor Inc has developed a device named "Coapsys" [97]. The Coapsys device uses a surgically implanted sub valvular chord to provide the tension to constrict the annulus [63]. To be accurate, the Coapsys device constricts not only the annulus, but also the region of the left ventricle associated with the papillary muscles [64]. The chord is shortened providing the tension which constricts the annulus and the associated region of the left ventricle [66]. The net result is a reduction in the A -P distance of the mitral valve. Kukamachi et al claim that this device is capable of producing a favourable alignment of both the papillary muscles and annulus through the tensioning effect of the subvalvular chord [97]. In affirmation of the results offered by Timek et al [73], and the results offered in this study, the Coapsys device reduces M R for type Il l-b dysfunction through annular constriction without any other surgical alterations to the mitral valve apparatus or leaflets [65]. One long term in-vivo animal study found the Coapsys device to reduce M R from a 68 graded 2.5 +/- 0.8 to 0.7 +/- 0.8 [65]. Another study found the M R to be reduced from a graded 2.9 +/- 0.7 to 0.6 +/- 0.7 [97]. These data support the hypothesis that it is possible to significantly reduce M R caused by type Il l-b dysfunction through annular constriction without any other surgical alteration of the valve. There are no published studies investigating the effectiveness of the Coapsys device for type II dysfunction. 4.3.3 Annular Constriction Using Undersized Rings Further evidence that annular constriction alone is capable of reducing M R for type Il l-b dysfunction can be found in clinical cases where the patient receives a prosthetic annular ring implant without any other surgical alteration of the valve. In cases of type Il l-b dysfunction where the leaflets are not diseased and the source of M R is cardiomyopathic dilation of the annulus, ventricle, and papillary muscles; constricting the annulus with a ring is typically effective in reducing M R clinically [98]. 4.4 Evidence Supporting the Importance of Annular Constriction for Type II Dysfunction To the best of our knowledge, this study is the first description of the effectiveness of annular constriction alone in reducing M R for type II dysfunction. Typical clinical surgery for type II dysfunction includes annuloplasty in conjunction with resection of the prolapsed leaflet and possible repair or replacement of chords [9, 48]. Since traditional open-heart procedures expose the chords and leaflets, traditional procedures can include direct repair of elongated or ruptured chords. Exposure of the chords and leaflets w i l l not be possible with certain newly emerging minimally invasive procedures such as percutaneous devices and the Coapsys device. It may be technically difficult or impossible for surgeons to surgically alter elongated or ruptured chords through minimally invasive techniques. The results of this study indicate that even i f the chords cannot be directly repaired, annular constriction alone is capable of reduce M R for type II dysfunction. 4.5 Why Does Annular Constriction Reduce Mitral Regurgitation? 4.5.1 Improvement of Leaflet Coaptation We believe there are two primary mechanisms reducing M R when the annulus is constricted. The first mechanism is related to the fact that annular constriction moves the 69 anterior and posterior leaflets closer together. During an in-vivo study of M R using ovine subjects, Tibayan et al confirmed that surgically reducing the A - P distance through annular constriction brought the posterior leaflet edge towards the anterior leaflet [72]. Timek et al believe the closer proximity of the anterior and posterior leaflets produced by septal-lateral cinching (annular constriction) improves leaflet coaptation [73]. Some have claimed the most significant reason why ring annuloplasty reduces M R is because the reduction in A - P distance facilitates improved coaptation between the leaflets [29, 73]. In a clinical study, echocardiography was used to measure the coaptation length of the anterior and posterior leaflets [99]. It was found that patients with M R have much lower coaptation lengths, (6.4 mm) compared to patients with normal or ring annuloplasty repaired hearts (11.0 and 11.6 respectively) [99]. The study shows that hearts repaired with ring annuloplasty are restored to near normal levels of leaflet coaptation length. Although not specifically measured during this study, it is likely that when the annulus was constricted, the coaptation length of the anterior and posterior leaflets was increased. M R was reduced because annular constriction improved leaflet coaptation. 4.5.2 Formation of a Monoleaflet Mitral Valve The other primary mechanism believed to contribute to the reduction of M R is the formation of a functionally monoleaflet valve. It has been recognized that ring annuloplasty procedures which use undersized rings can form monoleaflet valves where the posterior leaflet is functionally nullified and only the anterior leaflet opens and closes [100]. Green et al have reported that all rigid and flexible prosthetic rings impair the motion of the posterior leaflet [101]. After ring annuloplasty, the posterior leaflet is frozen in the open position with the anterior leaflet solely responsible for opening and closing the valve orifice [101]. It has been hypothesized that the formation of a monoleaflet valve may be less susceptible to regurgitation [29]. In a study o f annular cinching, Tibayan et al reported that when prosthetic rings were not used, and annular constriction was performed through tensioning of trans-annular sutures, the motion of the posterior leaflets was not impaired [71]. The modular device used in this study would likely produce similar results to those reported by Tibayan. Without a prosthetic ring the posterior leaflet would not be impaired in orientation or motion. However, constriction of the annulus would still have the effect of 70 reducing and eventually nullifying the functionality of the posterior leaflet. When the annulus was 50% constricted, qualitative observations affirm that the valve was essentially functioning as a monoleaflet. The annulus was constricted to the point that the closed anterior leaflet would be essentially coapting with the posterior annulus. For this study, regardless of whether it was a restricted, flail , or prolapsed leaflet, the simulation disease was always simulated on P2 of the posterior leaflet. It is logical that surgical intervention which nullifies the functionality of the diseased leaflet w i l l remove the symptoms of the disease. Simply put, when the annulus was constricted, the diseased posterior leaflet became less important and the healthy anterior leaflet became more important. A t a certain magnitude of constriction, the posterior leaflet became irrelevant as the anterior leaflet coapted with the posterior annulus; forming a competent monoleaflet valve. A s can be seen in Figures 3.6, 3.7, and 3.8; at 50% annular constriction, the essentially monoleaflet valve was measured to have very low regurgitation flow rates similar to baseline values for the flail, prolapsed, and restricted leaflet. It is important to note that even before the formation of a true monoleaflet; competency o f the valve could be restored as the annulus constricted enough to reduce the importance of the posterior leaflet to the point that the lesion was irrelevant. For example, from qualitative observations, annular constriction of 30% did not create a true monoleaflet. Nevertheless, the average regurgitation flow rate was measured to be less than a low 300 ml/min for flail , prolapsed, and restricted leaflets at 30% annular constriction. Based on the qualitative observations and the data produced in this study, it appears that when the annulus is constricted, the importance of the posterior leaflet is reduced. When a lesion is associated with the posterior leaflet, annular constriction produces a reduction in M R because the functionality and effect of the posterior leaflet is reduced or nullified depending on the magnitude o f annular constriction. 4.6 P o s s i b l e C h a n g e s i n the C h o r d a l F o r c e D i s t r i b u t i o n has a n U n k n o w n E f f e c t o n M R Chordal force distribution is a term to describe the balance of tension applied by the chords to the leaflets [17]. For a normal competent valve, the leaflets are suspended in a balanced configuration which enables coaptation during systole. Valvular lesions 71 disrupt the chordal force distribution creating imbalances which impair leaflet coaptation [25]. The chordal force distribution is a product of the interactions between all the components of the mitral valve apparatus. Alterations to any component o f the mitral valve apparatus can produce changes in the chordal force distribution because the mitral valve apparatus is a complex structure with tightly coupled components which dynamically interact with each other [102]. When there are geometric alterations to any component of the mitral valve apparatus, there w i l l be alterations in the chordal force distribution [102]. Timek et al have used radiopaque markers implanted on the tips of papillary muscles to show that annular constriction can change the location and orientation of the papillary muscles [96]. Constriction of the annulus causes movement of the papillary muscle tips towards the anterior annulus [96]. When the annulus is constricted, the entire mitral apparatus is affected and the papillary muscle locations wi l l be altered [102]. Correction of abnormal chordal force distribution may be another factor contributing to the reduction of M R when the annulus is constricted. Timek et al hypothesized that for ischemic M R with a type Il l-b restricted leaflet, septal-lateral cinching reduces M R through advantageous changes in subvalvular geometry [73]. This hypothesis is based on the studies that indicate annular constriction produces changes to the subvalvular apparatus and therefore changes to the chordal force distribution likely occur [72, 102]. The location and orientation of the papillary muscles were not measured during this study. Because of this, it is not possible to confirm that annular constriction produced changes in the location o f the papillary muscles for any of the created lesions. Also , the tension in each chord was not measured during this study and thus it is not known i f annular constriction produced changes to the chordal force distribution for any of the created lesions. Furthermore, i f a change in the chordal force distribution did occur when the annulus was constricted, it is impossible to know whether the changes were beneficial or detrimental for the reduction of M R . In theory, changes to the mitral valve apparatus created by annular constriction could create favourable or detrimental changes in the chordal force distribution. This is especially possible considering that the 72 three distinct lesions may react differently to changes in the chordal force distribution. The changes in chordal force distribution created by annular constriction may benefit one type of dysfunction but not another. Although studies have indicated that annular constriction can produce changes in the location of papillary muscles, and there have been studies which indicate that changes in the mitral valve apparatus directly correspond to changes in the chordal force distribution, this study cannot verify that annular constriction reduces M R through favourable changes in the chordal force distribution. It would be necessary to measure the chord tension in all the chords to monitor changes in chordal force distribution in order to test Timek's hypothesis. It is possible that changes in the chordal force distribution due to annular constriction play a small role in the reduction of M R . However, we believe that in this study, the two primary mechanisms of M R reduction (improved leaflet coaptation and formation of a monoleaflet valve) are more important than an effect produced by changes to the subvalvular apparatus due to annular constriction. 4.7 The Effect of Lesion Type 4.7.1 The Effect of Lesion Type on the Ability to Reduce Mitral Regurgitation through Annular Constriction The calculated regurgitation reduction percentage is useful because it enables comparison of the effectiveness to reduce M R without the need for direct referencing of the baseline regurgitation flow rates. The regurgitation reduction percentages for the three types of created lesions are presented in Figures 3.9, 3.10, and 3.11 for annular constrictions of 10%, 30%>, and 50%, respectively. Analysis of variance testing determined that the type of lesion prior to annuloplasty did not have a significant effect on how much annular constriction reduced regurgitation. A t all three magnitudes of annular constriction, there was no significant difference in regurgitation reduction percentage for flail , prolapsed, or restricted leaflets. We believe that the type of lesion prior to annuloplasty does not have a significant effect on the regurgitation reduction percentage because annular constriction reduces M R through the same mechanisms for all three lesions. The two primary reasons for a 73 reduction in M R are 1) improved leaflet coaptation 2) formation of a monoleaflet valve where the diseased posterior leaflet has reduced or no importance. A reduction in A - P distance w i l l produce an improved ability for leaflet coaptation for the three lesions studied. The restricted leaflet has improved coaptation because the anterior and posterior leaflets are closer together and therefore better able to make physical contact. For type II dysfunction, the closer proximity of the anterior and posterior leaflets has a direct effect on the ability of the posterior leaflet to prolapse. The anterior leaflet w i l l hold the posterior leaflet close to or against the posterior annulus (depending on the magnitude of annular constriction). Annular constriction improves coaptation of the leaflets for type II dysfunction because the posterior leaflet is unable to prolapse due to the interference of the anterior leaflet. The formation of a monoleaflet valve is a mechanism that has an equal effect on the reduction of M R for all o f the lesions examined in this study. When the annulus constricts and the importance of the posterior leaflet is reduced, the type of lesion wi l l not affect the reduction of regurgitation because the lesion becomes nullified. For example, at 50% annular constriction the anterior leaflet essentially coapts with the posterior annulus. The lesion on the posterior leaflet is nullified because the posterior leaflet no longer has a function for the valve. A t this level o f annular constriction all posterior leaflet lesions w i l l produce similar reductions in M R because the type o f lesion becomes irrelevant. 4.7.2 Limitations of the Calculated Regurgitation Reduction Percentage A t all three measured magnitudes of constriction, the average restricted leaflet has a lower regurgitation reduction percentage than the flail and prolapsed leaflet. Although there is not a statistically significant difference between the groups, there is a trend. It is believed that the restricted leaflet has a comparably lower average regurgitation reduction percentage because the restricted leaflet has a lower average initial disease regurgitation flow rate. The regurgitation reduction is calculated by dividing the annularly constricted regurgitation flow rate by the initial disease regurgitation flow rate. If the initial disease regurgitation flow rate is low, then the regurgitation reduction percentage w i l l be lower. For example, i f after annular constriction the regurgitation flow rate is 50 ml/min, the regurgitation reduction percentage w i l l be 97.5% i f the initial disease regurgitation flow 74 rate was 2000 ml/min, but only 90% i f the initial disease regurgitation flow rate was 500 ml/min. We believe that i f the restricted leaflet initial disease regurgitation flow rate was higher and closer to the rates seen with the prolapsed and flail leaflets, then the regurgitation reduction percentage would also be closer and there would not be a trend of lower values for the restricted leaflet. The trend is therefore a result of the method with which the regurgitation reduction percentage is calculated and is not necessarily representative of a difference in the reduction of M R for each type of lesion. 4.8 The Effect of Disease Severity on the Ability to Reduce Mitral Regurgitation through Annular Constriction From the data presented in Figure 3.12, the regurgitation flow rates at annular constrictions of 10% and 30% are affected by the initial severity of the disease. A t these magnitudes of constriction, when the initial disease regurgitation flow rate is greater, the regurgitation flow rate after annular constriction is also greater. Table 3.6 confirms that for annular constrictions of 10% and 30% there is a significant difference between the groups with different initial disease severities. The regurgitation flow rate after annular constriction is dependent on the initial disease flow rate. In contrast, the regurgitation flow rate after 50% annular constriction is independent of the initial disease flow rate. In Figure 3.12 the regurgitation flow rate at 50% annular constriction remains relatively constant regardless of the initial disease severity. A t all disease severities, 50% annular constriction reduced M R to essentially baseline levels. This is confirmed by Table 3.6 where a Welch A N O V A test indicates that at 50% annular constriction there is no significant difference in the regurgitation flow rate between groups with different initial disease severities. The initial disease severity has no effect on the regurgitation rate at 50%) annular constriction. If the goal of annuloplasty is to reduce M R to baseline values, then 30% or 50% annular constriction would be suitable i f the initial disease regurgitation flow rate is mild or moderate (1500 ml/min or less). If the initial disease regurgitation flow rate is severe (over 1500 ml/min); then only 50% annular constriction w i l l reduce M R to baseline values. However, i f the goal o f annuloplasty is to reduce M R to trivial or mild levels of M R , then 30% or 50% annular constriction would be suitable at all initial disease flow 75 rates including severe. A t 30% annular constriction severe disease is reduced to mild M R . 4.9 The Ideal Magnitude of Annular Constriction The device used during this study was not specifically tested to find the ideal constriction level. For all three lesions, 50% annular constriction produced the lowest average regurgitation flow rate. However, 30% annular constriction also produced very marked reductions in M R . The regurgitation flow rate at 30% constriction was often only slightly greater than the regurgitation flow rate at 50% constriction. In fact, for the three lesions tested, the data in Tables 3.3, 3.4, and 3.5 indicate that there were no significant differences between the regurgitation flow rates at 30% and 50% annular constriction. Annular constriction of 30%) may be the most ideal of the three levels tested in this study because the regurgitation flow rate is similar to the flow rate at 50% annular constriction with the advantage that less force is required to achieve 30% constriction. A t 50% annular constriction the modules must apply more downward force and the transverse sutures have greater tension. The effect on the myocardium of increased force applied by the device is unknown. Ideal annular constriction of approximately 30%> loosely agrees with a study of the Coapsys device where the ideal magnitude of annular constriction was investigated [63]. It was determined that the reduction of M R was maximized when the Coapsys device was constricted to 30% of its original length [63]. After 30% constriction there was no longer any reduction in M R [63]. It must be noted that for the Coapsys study, the measurements of constriction were measured from the subvalvular chord length and do not directly apply to the A - P distance. However, it is likely that the magnitude of device constriction would be correlated to the magnitude of annular constriction. 4.10 The Creation of Regurgitation through Valvular Lesions With a calculated p value of <0.0001, the Welch A N O V A test presented in Figure 3.5 verifies that the creation of mitral valve dysfunction produced regurgitation rates significantly different from the baseline regurgitation rate. Through the creation of specific lesions, type II and type Il l-b mitral valve dysfunctions were created. 76 The results of the Games-Howell test in Table 3.2 indicate that the flail leaflet, prolapsed leaflet, and restricted leaflet all had regurgitation flow rates greater than the baseline value. It was found that the type II dysfunctions created by chord elongation and chord cutting produced statistically similar levels of M R . The type Il l-b restricted leaflet created by shortened chords produced a lower level of M R . The regurgitation flow rate of the restricted leaflet was significantly lower than the flow rate of the prolapsed leaflet but not significantly different from the flow rate of the flail leaflet. Intuitively, it would not be unexpected for each disease group to have a significantly different level of M R . Since each group has a different lesion, it could be expected that the amount of M R would also differ. The reason why the lengthened chord and cut chord groups have similar regurgitation flow rates is likely a result of the similarity between the two lesions. Both lesions created type II dysfunction and therefore created similar magnitudes of M R . A s seen in Table 3.1, the average change in chord length was comparable between lengthened chords and shortened chords. Lengthened chords were elongated an average of 6.0 mm and shortened chords were shortened an average of 6.7 mm. Due to the lower average regurgitation flow rates of the restricted leaflet, it appears as though the posterior leaflet is less sensitive to chordal shortening compared to chord elongation or chord rupture. Therefore, in order to produce greater regurgitation flow rates for the restricted leaflet it is likely necessary to increase the magnitude of chordal shortening. A further study with a higher sample size of created lesions would be necessary to confirm this conclusion. In this study it was possible to reproducibly create mitral valve dysfunction through the simulation of three distinct lesions. A type II prolapsed leaflet was created by lengthening chords. A type II flail leaflet was simulated by cutting chords. A restricted type Ill-b leaflet was simulated by shortening chords. 4.11 Advantages of the Novel Model of Mitral Valve Disease 4.11.1 Real Hearts are used for the Creation of Mitral Regurgitation The in-vitro model developed and used in this study contains several beneficial features for the study of M R . One important feature of this model is that it uses a real heart as the platform for experimentation. This is in contrast with all other in-vitro models where a porcine valve is implanted onto an artificial ventricle with an artificial 77 annulus [74, 76, 77]. Although the other in-vitro models have carefully attempted to accurately produce the artificial left ventricle, we believe that using a true ventricle is beneficial. Through the use of a real heart, this model assures the anatomic and geometric arrangement o f the left ventricle and mitral valve apparatus are entirely correct before surgical manipulation and creation of the lesions. This belief is supported by the low baseline regurgitation flow rates. In all cases, the unaltered hearts had competent mitral valves with very low measured levels of regurgitation. Since it is known that the mitral valve apparatus commences the trial in its natural configuration and orientation, the amount of regurgitation created can be considered a direct result of the lesions created. The use of a real heart also has an important advantage during the experiment when the manipulation of the annulus is investigated. A s mentioned previously, the mitral valve apparatus is a complex structure with complex interactions between its components. The use of a real heart allows for observation of these complex interactions. The models which use an artificial ventricle and annulus are limited by the challenge of accurately simulating the mitral valve apparatus in all instances. It is impossible for a simplified, artificial model to unconditionally represent the complexity of a natural valve. 4.11.2 Visibility of the Mitral Valve The model used in this study offers the significant advantage of exposing the mitral valve for extraordinary visibility. It is possible to easily view the orientation o f the annulus and mitral leaflets. When a regurgitant jet is present, the magnitude of regurgitation can be qualitatively assessed through observation of the size of the jet. Also , through direct observation of the location and orientation of the jet, it is possible to determine the exact location of compromised coaptation. For example, i f the lateral portion of P2 is prolapsed above the plane of the anterior leaflet, this can be visually verified. 4.11.3 Accessibility of the Mitral Valve Related to the ease with which the mitral valve can be viewed, is the ease with which the mitral valve can be accessed and surgically manipulated. The model enables essentially unobstructed access to the entire mitral valve apparatus. It is possible to 78 evacuate and illuminate the ventricular lumen, exposing and granting easy access to the papillary muscles and all the chords. It is possible and practical to create lesions through the adjustment of the lengths of these chords with a precision of approximately one millimeter. Also , this model enables manipulation and measurement of the annulus within a precision of one millimeter. 4.12 Limitations of this Mitral Valve Disease Model 4.12.1 The Heart is Static and Lacks Hemodynamically Accurate Fluid Flow Although our model worked well for this study; there are some notable limitations of the model. The most significant limitation is the obvious fact that this model is not dynamic. Unlike an in-vivo model where the heart undergoes the normal cardiac cycle with the valve opening and closing, this in-vitro model can only investigate a static ventricle and leaflets. The ventricle is continuously pressurized and the heart is subjected to a steady state environment for the duration of the trial. The disadvantage of this steady state is that it is impossible to study how the opening and closing o f the valve would affect regurgitation. This model assumes that the opening and closing of the valve would not affect the regurgitation during mid systole. Another significant disadvantage of a static heart model is that it is impossible to incorporate the natural changes in the shape of the heart when the cardiac muscle contracts during systole. For each cycle, beating hearts undergo natural changes in the shape and orientation of the ventricle, annulus, and papillary muscles. This model is unable to predict the effects of these conformation changes during systole. Unlike the sophisticated in-vitro model developed at Georgia Tech, the model used in this study does not attempt to simulate hemodynamically accurate fluid flow through the ventricle. In fact, unless there is regurgitation, this model does not have any fluid flow through the ventricle. The ventricle is pressurized to a physiologically relevant level and all fluid flow is retrograde regurgitation across the mitral valve into the atrium. A t no point does fluid ever travel antegrade from the ventricle into the aorta. 4.12.2 Comparison with Clinically Measured Mitral Regurgitation Due to the fact that the model used in this study is not dynamic, it is impossible to make direct comparisons between the levels of M R found in this model and the levels of 79 M R measured clinically. The circumstances of regurgitation are too different to enable quantitative comparison. With the model, the heart is constantly pressurized and therefore regurgitation is taking place for the entire five minute trial. Clinically, during a five minute observation period, the heart w i l l only be in systole for roughly half of the time. Furthermore, regurgitation does not occur uniformly during systole [103]. During the cardiac cycle, M R wi l l only occur during a fraction of the total cycle time. For this reason quantitative comparison between model regurgitation and clinical regurgitation cannot be performed. 4.12.3 Absence of Atrial Pressure One minor limitation of this in-vitro model is the absence o f any significant atrial pressure. The left ventricle is pressurized, but the left atrium is removed in order to expose the mitral valve. The trans-mitral pressure is therefore a product of the left ventricular pressure and the essentially inconsequential air pressure on the outside of the mitral valve. In the absence of atrial pressure, one might expect deformation of the leaflets. During this study, it did not appear as though deformation of the leaflets occurred due to removal of the atrium. In evidence of this, one can review the baseline levels of M R at the commencement of each trial. In all cases the hearts had very low, almost imperceptible levels of M R before the simulation of disease. In fact, no baseline hearts had visible regurgitant jets of any size before disease simulation. The absence of a pressurized atrium did not appear to affect the orientation of the leaflets and therefore the level of M R in this study. This is likely due to the fact that the ventricle was pressurized to 107 mmHg and the normal pressure of the porcine left atrium has been measured to be only 5 mmHg [104]. In both normal hearts and this model, the left ventricular pressure always far exceeds the left atrial pressure. 4.12.4 Healthy Hearts are used to Simulate Disease Another limitation of this in-vitro model is that the hearts used are from healthy animals. There are no infarctions, scar tissue, or ventricular remodeling. The shape and orientation of the ventricle, papillary muscles, and annulus are in a natural, healthy configuration. Although the created lesions effectively created M R , we must not forget 80 that the tissue is not diseased and it is possible that truly diseased hearts may react to surgical manipulation in a subtly or perhaps significantly different manner. 4.13 Annuloplasty Using the Novel Device 4.13.1 Analysis of Changes in the Anterior-Posterior Distance The ability of the device to effectively and precisely constrict the annulus is evident in Figure 3.1. During the study, the magnitude of annular constriction was determined by the magnitude of reduction of the A - P distance. Therefore, the A - P distance is a fixed variable. The standard error is relatively small because the device was capable of precise constriction and also because the ten hearts used in the study were of similar size with similar baseline A - P distances. 4.13.2 Analysis of the Annulus Area When the device is tensioned and the A - P distance is reduced, Figure 3.2 shows that the annulus area is reduced. The mean reduction in annulus area (%) closely resembles the magnitude of A - P distance reduction. For example, when the A - P distance is reduced to 50% of its baseline value, the calculated annulus area is 51%» of its baseline value. The reason the annulus area changes in proportion to the A - P distance is because there is very little measured change in the distance along the anterior annulus. In all cases, the long axis from trigone to trigone does not change more than 2.0 mm. The most significant change to the semi-elliptical annulus is the short axis; the A - P distance. Thus, when the long axis remains near the same, and the short axis is reduced by half, the total area w i l l be essentially reduced by half as well . 4.13.3 Comparison of Annulus Area Reduction The magnitude of annulus area reduction is comparable to the reduction seen in clinical trials using commonly accepted annuloplasty techniques. The use of annuloplasty rings is reported to commonly produce annulus area reductions of approximately 30% [73]. O f course, the amount of area reduction w i l l vary greatly depending on the size of ring chosen by the surgeon. Also , flexible rings have been observed to allow for changes in the annulus area to occur during the normal cardiac cycle. The changes in annulus area during a normal cardiac cycle have been reported in the range o f 10-28% depending on the study and type of ring used [105]. After ring 81 annuloplasty, acceptable mean annulus areas range from approximately 200 to 500 mm . Clinical studies have documented successful ring annuloplasties to have measured annulus areas of: 280, 340, 210, 460, 480, 421, and 380 m m 2 [106-110]. The variation in annulus area would be a result of many factors. Some of the most significant factors would be: individual surgeon preference for a specific brand or size of prosthetic rings, patient heart size, and patient pathology. 4.13.4 Stenosis after Annuloplasty In general, annulus areas greater than 150 m m 2 are considered to be acceptable [107]. Patients with lesser annulus areas are likely to become stenotic and symptomatic. In a published case study, post-annuloplasty patients with a mean annulus area o f 85 mm were symptomatic and required valve replacement due to mitral valve stenosis [111]. The annulus areas calculated throughout this study are similar to the areas created using conventional annuloplasty techniques. A t 50% constriction, the average annulus area remained an acceptable 332 mm 2 . Stenosis would be an uncommon problem at this constriction level. 4.13.5 Analysis of Annulus Circumference Figure 3.3 depicts the effect of annular constriction on the annulus circumference. When the magnitude of annular constriction is greater, there is a greater reduction in annulus circumference. A reduction in annulus circumference is important to consider because it verifies that actual constriction of the annulus has occurred. Theoretically, it could be possible to reduce the A - P distance without changing the circumference of the annulus. If the device only created a change of shape of the annulus, then there would be no change in the annulus circumference. The changes in annulus circumference verify that the device not only alters the annulus shape, it also constricts the annulus. The reduction in annulus area and A - P distance is a result of changes in the annulus shape as well as a reduction in circumference. 82 4.14 Conformational Changes in Sections along the Annulus 4.14.1 Changes in the Distances between Markers along the Anterior Annulus Presented in Figure 3.4, the data from the measured distances between the annulus markers reveals three noticeable trends. The first trend is that the distance between markers 1 and 3 did not change very much. These markers span the anterior annulus. It can be concluded that manipulation and constriction of the posterior annulus does not greatly affect this inter-trigonal distance. This result is not surprising when the anatomical configuration of the anterior annulus is considered. The anterior annulus is composed of fibrous tissue in close association with the pressurized aorta. The fibrous anterior section of the annulus is reported to be relatively rigid and less susceptible to dilation [105]. However, it does have subtle dynamic properties and it has been reported to change in size in response to hemodynamic loading and ventricular contractility [112]. The magnitude in change across the anterior annulus is relatively small compared to the rest of the annulus. This study has verified that surgical manipulation of the posterior annulus is capable of reducing M R without creating large changes to the anterior annulus. 4.14.2 Changes in the Distances between Markers along the Posterior Annulus The second trend in Figure 3.11 is that magnitude of constriction is similar between the markers 3 to 4, 4 to 5, 5 to 6, and 6 to 1. The sections between these markers cover the entire circumference o f the annulus except for the anterior section already discussed. From these results it appears as though the device constricted the posterior section of the annulus relatively equally. Notably, the values for the constriction of section 4 to 5 and section 5 to 6 are quite similar throughout the study. These two sections span the posterior annulus from P I , through P2, and into P3. Equal and balanced constriction especially across this area is believed to be beneficial because it maintains a favourable orientation of the posterior leaflet. Consider for example i f there was a gross imbalance to this posterior section of the annulus. The posterior leaflet would become misaligned, impairing coaptation with the anterior leaflet. A qualitative observation made during the pre-trial developmental stages of this study found that i f the annulus was constricted in a grossly asymmetrical manner, M R would not be effectively reduced. For this reason, the device was carefully placed across the posterior annulus in an attempt to create a balanced and equal constriction. 83 It is apparent that there is a slight difference in the constriction levels for sections 3 to 4 and 6 to 1. If the posterior annulus was moved towards the anterior annulus in a perfectly symmetrical manner, one would expect the constriction values for these sections to be identical. The slight variation is believed to be a result of minor miscalculations in the placement of the device. Although a perfectly symmetrical constriction would be preferred, the slight variation measured did not appear to affect the ability of annular constriction to reduce M R . After gaining the experience of implanting the device throughout this study, it is believed that during a subsequent study it would be possible to improve the symmetry of constriction. Perhaps a small improvement in symmetry may further improve the ability to reduce M R . 4.14.3 Changes in the Distances between Markers Spanning the Valve Orifice The third trend noticeable in Figure 3.11 is that the values of constriction for sections 2 to 4, 2 to 5, and 2 to 6 are typically similar and of the greatest magnitude. It is not surprising that these sections have the greatest measured constriction because these are the measurements spanning across the valve orifice. A s mentioned, the distance between markers 2 and 5 is the A - P distance. The values for sections 2 to 5 and 2 to 6 are very similar for the entire study. The values for section 2 to 4 are consistently lower in all cases. This indicates another example of asymmetry in the constriction. A s before, this magnitude of asymmetry is considered minimal. It is not known whether it hindered the ability of constriction to reduce M R . Improvement in the symmetry of constriction in this area may also further increase the ability of annular constriction to reduce M R . 84 Chapter 5 - Conclusion 5.1 Annular Constriction Reduces Mitral Regurgitation The results of this study strongly indicate that when the M R is due to type II or type Il l-b dysfunction affecting the posterior leaflet, it is possible to reduce M R through annular constriction alone. In the absence of any other surgical manipulation to the valve, annular constriction reduced M R for a flail leaflet due to ruptured chords, a prolapsed leaflet due to lengthened chords, and a restricted leaflet due to shortened chords. Annular constriction reduces M R through two primary mechanisms. Annular constriction improves leaflet coaptation by bringing the anterior and posterior leaflets into closer proximity. Improved leaflet coaptation results in improved valve competency. The second important mechanism is the formation of a monoleaflet valve. When the annulus constricts, the importance of the posterior leaflet is reduced and the effect of the lesions associated with the posterior leaflet are reduced as well . Annular constrictions of 50% found the anterior leaflet essentially coapting with the posterior annulus. A t this magnitude of annular constriction the valve has formed a monoleaflet and the posterior leaflet is nonfunctional. A n y lesions associated with the posterior leaflet become irrelevant and valve competency is restored. 5.2 The Lesion Type does not affect the Reduction of Regurgitation due to Annular Constriction The flail, prolapsed, and restricted leaflet all responded to annular constriction with similar reductions in M R . For each magnitude of annular constriction, no significant difference in the regurgitation reduction was found between the different lesions. The reason both type II and type Il l-b dysfunctions have similar reductions in M R for each magnitude of annular constriction is because the mechanisms of M R reduction apply similarly for all the created lesions. For the restricted, prolapsed, and flail leaflets, annular constriction improves leaflet coaptation and forms a monoleaflet valve thereby reducing M R . 85 5.3 The Initial Disease Severity Affects the Regurgitation Rate at 10% and 30% but not 50% Annular Constriction For initial disease severity of mild or moderate, both 30% and 50% annular constriction reduced M R to essentially baseline levels. For initially severe disease, only 50%) annular constriction reduced M R to essentially baseline levels. The regurgitation flow rate at annular constrictions of 10%> and 30% was found to be affected by the initial disease severity. The regurgitation flow rate at 50% annular constriction was not affected by the initial disease severity. 5.4 Alteration of the Chords Created Flail, Prolapsed, and Restricted Posterior Leaflets The porcine heart in-vitro model effectively produced type II and type Il l-b mitral regurgitation through surgical alteration of the chords. A flail posterior leaflet was created by cutting chords; a prolapsed posterior leaflet was created by lengthened chords; and a restricted posterior leaflet was created by shortening chords. The heart cage and device developed for this study allowed for precise manipulation of the mitral annulus and measurement of mitral regurgitation. These apparatuses would be well suited for further investigations of mitral valve disease and surgical annuloplasty. 5.5 Avenues of Future Study Comparison of clinical pathology with the in-vitro model is limited because the porcine hearts had normal sized annuli, histologically normal leaflets, and normal ventricular conformations. In order to further study the effects of annular constriction, it would be beneficial to perform an in-vivo animal study with clinically relevant valvular lesions. Echocardiography would be ideal to monitor M R as well as conformational changes in the mitral valve apparatus produced by constriction of the annulus. The dynamic environment of the beating heart would produce clinically relevant data and further test the hypothesis that annular constriction without any other surgical manipulation is capable of reducing mitral regurgitation. 86 References 1. Barlow, J.B. , Perspectives on the Mitral Valve. 1987, Philadelphia: F . A . Davis Company. 2. Cheitlin, M . D . and R . C . Byrd, Mitral Valve Disease, in Valvular Heart Disease, B . H . Greenberg and E . Murphy, Editors. 1987, P S G Publishing Company, Inc: Littleton, Massachusetts, p. 92-122. 3. Frankel, S.K., L .S . L i l l y , and J .A. Bit t l , Valvular Heart Disease, in Pathophysiology of Heart Disease, L . S . L i l l y , Editor. 1998, Will iams & Wilkins: Baltimore, Maryland, p. 171-192. 4. Smith, D.R. , Clinical Diagnosis and Evaluation of Mitral Valve Disease, in Mitral Valve Disease, M . I . Ionescu and L . H . Cohn, Editors. 1985, Butterworth & Co. Ltd: Toronto, p. 43-51. 5. Frater, R . W . M . , Functional Anatomy of the Mitral Valve, in Mitral Valve Disease, M . I . Ionescu and L . H . Cohn, Editors. 1985, Butterworth & Co. Ltd : Toronto, p. 127-138. 6. Jeresaty, R . M . , Mitral Valve Prolapse. 1979, N e w York: Raven Press. 7. Rankin, J.S., M . S . J . Hickey, L . R . Smith, N . P . DeBruijn, K . H . Sheikh, D . C . Sabiston, Jr., Current concepts in the pathogenesis and treatment of ischemic mitral regurgitation, in Ischemic Mitral Incompetence, R . H . H . O . Vetter, H . Schmutzler, Editor. 1988, Steinkopff Verlag Darmstadt: N e w York. p. 157-178. 8. Haffajee, C.I., Chronic Mitral Regurgitation, in Valvular Heart Disease, J .E. Dalen and J.S. Alpert, Editors. 1987, Little, Brown and Company: Toronto, p. 111-150. 9. Carpentier, A . , Cardiac valve surgery—the "French correction". J Thorac Cardiovasc Surg, 1983. 86(3): p. 323-37. 10. Reece, T .B. , et al., Mitral repair is superior to replacement when associated with coronary artery disease. A n n Surg, 2004. 239(5): p. 671-5; discussion 675-7. 11. Cohn, L . H . , et al., Minimally invasive cardiac valve surgery improves patient satisfaction while reducing costs of cardiac valve replacement and repair. A n n Surg, 1997. 226(4): p. 421-6; discussion 427-8. 12. Devereux, R . B . , R. Kramer-Fox, and P. Kligfield, Mitral valve prolapse: causes, clinical manifestations, and management. A n n Intern Med, 1989.111(4): p. 305-17. 13. Bouknight, D.P. and R . A . O'Rourke, Current management of mitral valve prolapse. A m Fam Physician, 2000. 61(11): p. 3343-50, 3353-4. 14. Hayek, E . , C . N . Gring, and B .P . Griffin, Mitral valve prolapse. Lancet, 2005. 365(9458): p. 507-18. 15. Barber, J.E., et al., Myxomatous mitral valve chordae. I: Mechanical properties. J Heart Valve Dis, 2001.10(3): p. 320-4. 16. Barber, J.E., et al., Mechanical properties of myxomatous mitral valves. J Thorac Cardiovasc Surg, 2001.122(5): p. 955-62. 87 17. Nielsen, S.L., et al., Imbalanced chordal force distribution causes acute ischemic mitral regurgitation: mechanistic insights from chordae tendineae force measurements in pigs. J Thorac Cardiovasc Surg, 2005. 129(3): p. 525-31. 18. Roman, M . J . , et al., Comparison of cardiovascular and skeletal features of primary mitral valve prolapse and Marfan syndrome. A m J Cardiol, 1989. 63(5): p. 317-21. 19. L ing , L . H . , et al., Clinical outcome of mitral regurgitation due to flail leaflet. N Engl J Med , 1996. 335(19): p. 1417-23. 20. M i l l s , W.R. , et al., Biomechanical and echocardiographic characterization of flail mitral leaflet due to myxomatous disease: further evidence for early surgical intervention. A m Heart J , 2004. 148(1): p. 144-50. 21. Roberts, W . C . , C . L . Mcintosh, and R . B . Wallace, Mechanisms of severe mitral regurgitation in mitral valve prolapse determined from analysis of operatively excised valves. A m Heart J, 1987.113(5): p. 1316-23. 22. Vassiliades, T .A . , Jr., et al., The clinical development of percutaneous heart valve technology: a position statement of the Society of Thoracic Surgeons (STS), the American Association for Thoracic Surgery (AATS), and the Society for Cardiovascular Angiography and Interventions (SCAI) Endorsed by the American College of Cardiology Foundation (ACCF) and the American Heart Association (AHA). J A m C o l l Cardiol, 2005. 45(9): p. 1554-60. 23. Nielsen, S.L., et al., Papillary muscle misalignment causes multiple mitral regurgitant jets: an ambiguous mechanism for functional mitral regurgitation. J Heart Valve Dis, 1999. 8(5): p. 551-64. 24. He, S., et al., Mitral valve compensation for annular dilatation: in vitro study into the mechanisms of functional mitral regurgitation with an adjustable annulus model. J Heart Valve Dis , 1999. 8(3): p. 294-302. 25. Nielsen, S.L., et al., Chordal force distribution determines systolic mitral leaflet configuration and severity of functional mitral regurgitation. J A m C o l l Cardiol, 1999. 33(3): p. 843-53. 26. He, S., et al., Integrated mechanism for functional mitral regurgitation: leaflet restriction versus coapting force: in vitro studies. Circulation, 1997. 96(6): p. 1826-34. 27. Nielsen, S.L., et al., Mechanism of incomplete mitral leaflet coaptation— interaction of chordal restraint and changes in mitral leaflet coaptation geometry. Insight from in vitro validation of the premise offorce equilibrium. J Biomech Eng, 2002.124(5): p. 596-608. 28. Otsuji, Y . , et al., Restricted diastolic opening of the mitral leaflets in patients with left ventricular dysfunction: evidence for increased valve tethering. J A m C o l l Cardiol, 1998. 32(2): p. 398-404. 29. L a i , D.T. , et al., The effects of ring annuloplasty on mitral leaflet geometry during acute left ventricular ischemia. J Thorac Cardiovasc Surg, 2000. 120(5): p. 966-75. 30. Gorman, R .C . , et al., Pathogenesis of acute ischemic mitral regurgitation in three dimensions. J Thorac Cardiovasc Surg, 1995.109(4): p. 684-93. 88 31. Gorlin, R., Natural History, Medical Therapy and Indications for Surgery in Mitral Valve Disease, in Mitral Valve Disease, M . I . Ionescu and L . H . Cohn, Editors. 1985, Butterworth & Co. Ltd: Toronto, p. 105-123. 32. Kalkanis, S.N. , et al., Cardiovascular Drugs, in Pathophysiology of Heart Disease, L . S . L i l l y , Editor. 1998, Will iams & Wilkins: Baltimore, p. 347-392. 33. Savage, E . B . , T .B . Ferguson, Jr., and V . J . DiSesa, Use of mitral valve repair: analysis of contemporary United States experience reported to the Society of Thoracic Surgeons National Cardiac Database. A n n Thorac Surg, 2003. 75(3): p. 820-5. 34. Ionescu, M . I . and M . Chidambaram, Operative Techniques for Mitral Valve Replacement, in Mitral Valve Disease, M . I . Ionescu and L . H . Cohn, Editors. 1985, Butterworth & Co. Ltd: Toronto, p. 191-205. 35. Hung, J . , et al., Reverse ventricular remodeling reduces ischemic mitral regurgitation: echo-guided device application in the beating heart. Circulation, 2002.106(20): p. 2594-600. 36. Walkes, J .C. and M . J . Reardon, Status of mitral valve surgery. Curr Opin Cardiol, 2004.19(2): p. 117-22. 37. Duran, C . M . G . , Conservative Surgery of the Mitral Valve - Ring Annuloplasties, in Mitral Valve Disease, M . I . Ionescu and L . H . Cohn, Editors. 1985, Butterworth & Co. Ltd: Toronto, p. 179-188. 38. Carpentier, A . , [Reconstructive valvuloplasty. A new technique of mitral valvuloplasty]. Presse Med, 1969. 77(7): p. 251-3. 39. Raffoul, R., et al., Clinical evaluation of the Physio annuloplasty ring. Chest, 1998.113(5): p. 1296-301. 40. Tahta, S.A., et al., Outcome after mitral valve repair for functional ischemic mitral regurgitation. J Heart Valve Dis, 2002.11(1): p. 11-8; discussion 18-9. 41. Matsunaga, A . , S.A. Tahta, and C M . Duran, Failure of reduction annuloplasty for functional ischemic mitral regurgitation. J Heart Valve Dis , 2004. 13(3): p. 390-7; discussion 397-8. 42. Grossi, E . A . , et al., Late results of isolated mitral annuloplasty for "functional" ischemic mitral insufficiency. J Card Surg, 2001. 16(4): p. 328-32. 43. Mesana, T., M . Ibrahim, and M . Hynes, A technique for annular plication to facilitate sliding plasty after extensive mitral valve posterior leaflet resection. A n n Thorac Surg, 2005. 79(2): p. 720-2. 44. Grossi, E . A . , et al., Early results ofposterior leaflet folding plasty for mitral valve reconstruction. A n n Thorac Surg, 1998. 65(4): p. 1057-9. 45. Gil l inov, A . M . and D . M . Cosgrove, 3rd, Modified quadrangular resection for mitral valve repair. A n n Thorac Surg, 2001. 72(6): p. 2153-4. 46. Fasol, R. and K . Mahdjoobian, Repair of mitral valve billowing and prolapse (Barlow): the surgical technique. A n n Thorac Surg, 2002. 74(2): p. 602-5. 47. Pomerantzeff, P . M . , et al., Mitral valve repair. Quadrangular resection of the posterior leaflet in patients with myxomatous degeneration. A r q Bras Cardiol, 1999. 73(3): p. 277-280. 48. Gil l inov, A . M . , et al., Durability of mitral valve repair for degenerative disease. J Thorac Cardiovasc Surg, 1998.116(5): p. 734-43. 89 49. Sakamoto, Y . , et al., Long-term assessment of mitral valve reconstruction with resection of the leaflets: triangular and quadrangular resection. A n n Thorac Surg, 2005. 79(2): p. 475-9. 50. Braunberger, E . , et al., Very long-term results (more than 20 years) of valve repair with carpentier's techniques in nonrheumatic mitral valve insufficiency. Circulation, 2001.104(12 Suppl 1): p. 18-11. 51. Grossi, E . A . , et al., Anterior leaflet procedures during mitral valve repair do not adversely influence long-term outcome. J A m C o l l Cardiol, 1995. 25(1): p. 134-6. 52. Alf ier i , O., et al., "Edge-to-edge" repair for anterior mitral leaflet prolapse. Semin Thorac Cardiovasc Surg, 2004. 16(2): p. 182-7. 53. Alf ier i , O., et al., The double-orifice technique in mitral valve repair: a simple solution for complex problems. J Thorac Cardiovasc Surg, 2001.122(4): p. 674-81. 54. Timek, T .A . , et al., Edge-to-edge mitral valve repair without ring annuloplasty for acute ischemic mitral regurgitation. Circulation, 2003. 108 Suppl 1: p. I l l 22 -7. 55. Mil ler , D . C . , Ischemic mitral regurgitation redux—to repair or to replace? J Thorac Cardiovasc Surg, 2001.122(6): p. 1059-62. 56. Levine, R . A . , et al., Mechanistic insights into functional mitral regurgitation. Curr Cardiol Rep, 2002. 4(2): p. 125-9. 57. Gorman, J .H. , 3rd, et al., Papillary muscle discoordination rather than increased annular area facilitates mitral regurgitation after acute posterior myocardial infarction. Circulation, 1997. 96(9 Suppl): p. II-124-7. 58. Menicanti, L . , et al., Ischemic mitral regurgitation: intraventricular papillary muscle imbrication without mitral ring during left ventricular restoration. J Thorac Cardiovasc Surg, 2002. 123(6): p. 1041-50. 59. Liel-Cohen, N . , et al., Design of a new surgical approach for ventricular remodeling to relieve ischemic mitral regurgitation: insights from 3-dimensional echocardiography. Circulation, 2000.101(23): p. 2756-63. 60. Downing, S.W., et al., Beating-heart mitral valve surgery: preliminary model and methodology. J Thorac Cardiovasc Surg, 2002.123(6): p. 1141-6. 61. Alf ier i , O., et al., Novel suture device for beating-heart mitral leaflet approximation. A n n Thorac Surg, 2002. 74(5): p. 1488-93. 62. Farm, J.I., et al., Beating heart catheter-based edge-to-edge mitral valve procedure in a porcine model: efficacy and healing response. Circulation, 2004. 110(8): p. 988-93. 63. Fukamachi, K . , et al., Optimal mitral annular andsubvalvular shape change created by the Coapsys device to treat functional mitral regurgitation. Asaio J, 2005. 51(1): p. 17-21. 64. Fukamachi, K . , et al., Changes in mitral annular and left ventricular dimensions and left ventricular pressure-volume relations after off-pump treatment of mitral regurgitation with the Coapsys device. Eur J Cardiothorac Surg, 2004. 25(3): p. 352-7. 65. Inoue, M . , et al., The Coapsys device to treat functional mitral regurgitation: in vivo long-term canine study. J Thorac Cardiovasc Surg, 2004. 127(4): p. 1068-76; discussion 1076-7. 90 66. Fukamachi, K . , et al., Reduction of mitral regurgitation using the Coapsys device: a novel ex vivo method using excised recipients' hearts. Asaio J, 2005. 51(1): p. 82-4. 67. Block, P .C. , Percutaneous mitral valve repair: are they changing the guard? Circulation, 2005.111(17): p. 2154-6. 68. Maniu, C . V . , et al., Acute and chronic reduction of functional mitral regurgitation in experimental heart failure by percutaneous mitral annuloplasty. J A m C o l l Cardiol, 2004. 44(8): p. 1652-61. 69. Liddicoat, J.R., et al., Percutaneous mitral valve repair: a feasibility study in an ovine model of acute ischemic mitral regurgitation. Catheter Cardiovasc Interv, 2003. 60(3): p. 410-6. 70. Daimon, M . , et al., Percutaneous mitral valve repair for chronic ischemic mitral regurgitation: a real-time three-dimensional echocardiography study in an ovine model. Circulation, 2005.111(17): p. 2183-9. 71. Tibayan, F . A . , et al., Does septal-lateral annular cinching work for chronic ischemic mitral regurgitation? J Thorac Cardiovasc Surg, 2004. 127(3): p. 654-63. 72. Tibayan, F . A . , et al., Annular or subvalvular approach to chronic ischemic mitral regurgitation? J Thorac Cardiovasc Surg, 2005.129(6): p. 1266-75. 73. Timek, T .A . , et al., Septal-lateral annular cinching abolishes acute ischemic mitral regurgitation. J Thorac Cardiovasc Surg, 2002. 123(5): p. 881-8. 74. Hopmeyer, J., et al., Estimation of mitral regurgitation with a hemielliptic curve-fitting algorithm: in vitro experiments with native mitral valves. J A m Soc Echocardiogr, 1998.11(4): p. 322-31. 75. Jimenez, J .H. , et al., Effects of a saddle shaped annulus on mitral valve function and chordal force distribution: an in vitro study. A n n Biomed Eng, 2003. 31(10): p. 1171-81. 76. Arita, M . , et al., Multiple purpose simulator using a natural porcine mitral valve. Asian Cardiovasc Thorac Ann , 2004.12(4): p. 350-6. 77. Lefebvre, X . P . , A . P . Yoganathan, and R . A . Levine, Insights from in-vitro flow visualization into the mechanism of systolic anterior motion of the mitral valve in hypertrophic cardiomyopathy under steady flow conditions. J Biomech Eng, 1992.114(3): p. 406-13. 78. Kunzelman, K . S . , et al., Finite element analysis of the mitral valve. J Heart Valve Dis, 1993. 2(3): p. 326-40. 79. Cochran, R.P . and K . S . Kunzelman, Effect of papillary muscle position on mitral valve function: relationship to homografts. A n n Thorac Surg, 1998. 66(6 Suppl): p. S155-61. 80. Kunzelman, K . S . , M . S . Reimink, and R.P. Cochran, Annular dilatation increases stress in the mitral valve and delays coaptation: a finite element computer model. Cardiovasc Surg, 1997. 5(4): p. 427-34. 81. Kunzelman, K . S . , et al., Acute mitral valve regurgitation created in sheep using echocardiographic guidance. J Heart Valve Dis, 1999. 8(6): p. 637-43. 82. Hennein, H . A . , et al., Left ventricular function in experimental mitral regurgitation with intact chordae tendineae. J Thorac Cardiovasc Surg, 1993. 105(4): p. 624-32. 91 83. Timek, T .A . , et al., Pathogenesis of mitral regurgitation in tachycardia-induced cardiomyopathy. Circulation, 2001.104(12 Suppl 1): p. 147-53. 84. Llaneras, M . R . , et al., Pathogenesis of ischemic mitral insufficiency. J Thorac Cardiovasc Surg, 1993.105(3): p. 439-42; discussion 442-3. 85. Timek, T .A . , et al., Will a partial posterior annuloplasty ring prevent acute ischemic mitral regurgitation? Circulation, 2002.106(12 Suppl 1): p. 133-139. 86. Llaneras, M . R . , et al., Large animal model of ischemic mitral regurgitation. A n n Thorac Surg, 1994. 57(2): p. 432-9. 87. Tibayan, F . A . , et al., Annular remodeling in chronic ischemic mitral regurgitation: ring selection implications. A n n Thorac Surg, 2003. 76(5): p. 1549-54; discussion 1554-5. 88. Jensen, M . O . , A . A . Fontaine, and A . P . Yoganathan, Improved in vitro quantification of the force exerted by the papillary muscle on the left ventricular wall: three-dimensional force vector measurement system. A n n Biomed Eng, 2001.29(5): p. 406-13. 89. Schwammenthal, E . , et al., Mechanism of mitral regurgitation in hypertrophic cardiomyopathy: mismatch of posterior to anterior leaflet length and mobility. Circulation, 1998. 98(9): p. 856-65. 90. Spence, A . P . and E . B . Mason, Human Anatomy and Physiology. 4th ed. 1992, St Paul, M N : West Publishing Company. 91. Whitlock, M . and D . Schluter, Introduction to Biostatistics. 1999, Vancouver: University of British Columbia. 92. Sail, J., A . Lehman, and L . Creighton, JMP Start Statistics. 2nd ed. 2001, Pacific Grove, C A : Duxbury. 93. Zar, J .H. , Biostatistical Analysis. 4th ed. 1999, Upper Saddle River, N e w Jersey: Prentice-Hall, Inc. 94. Toothacker, L . E . , Multiple Comparisons Procedures. Quantitative applications in the social sciences. V o l . 89. 1993, Newbury Park, C A : Sage Publications. 95. Timek, T .A . , et al., Effects of paracommissural septal-lateral annular cinching on acute ischemic mitral regurgitation. Circulation, 2004.110(11 Suppl 1): p. 1179-84. 96. Timek, T .A . , et al., Annular versus subvalvular approaches to acute ischemic mitral regurgitation. Circulation, 2002.106(12 Suppl 1): p. 127-132. 97. Fukamachi, K . , et al., Off-pump mitral valve repair using the Coapsys device: a pilot study in a pacing-induced mitral regurgitation model. A n n Thorac Surg, 2004. 77(2): p. 688-92; discussion 692-3. 98. Accola , K . D . , et al., Midterm outcomes using the physio ring in mitral valve reconstruction: experience in 492patients. A n n Thorac Surg, 2005. 79(4): p. 1276-83; discussion 1276-83. 99. Yamauchi, T., et al., Evaluation of the mitral valve leaflet morphology after mitral valve reconstruction with a concept "coaptation length index". J Card Surg, 2004.19(6): p. 535-8. 100. Dobre, M . , B . K o u l , and A . Rojer, Anatomic and physiologic correction of the restricted posterior mitral leaflet motion in chronic ischemic mitral regurgitation. J Thorac Cardiovasc Surg, 2000.120(2): p. 409-11. 92 101. Green, G.R., et al., Restricted posterior leaflet motion after mitral ring annuloplasty. A n n Thorac Surg, 1999. 68(6): p. 2100-6. 102. Dagum, P., et al., Coordinate-free analysis of mitral valve dynamics in normal and ischemic hearts. Circulation, 2000.102(19 Suppl 3): p. III62-9. 103. Enriquez-Sarano, M . , et al., Changes in effective regurgitant orifice throughout systole in patients with mitral valve prolapse. A clinical study using the proximal isovelocity surface area method. Circulation, 1995. 92(10): p. 2951-8. 104. Szabo, G . , et al., A new computer model of mitral valve hemodynamics during ventricular filling. Eur J Cardiothorac Surg, 2004. 26(2): p. 239-47. 105. Dagum, P., et al., Three-dimensional geometric comparison of partial and complete flexible mitral annuloplasty rings. J Thorac Cardiovasc Surg, 2001. 122(4): p. 665-73. 106. Camjlleri, L . F . , et al., Flexible posterior mitral annuloplasty: five-year clinical and Doppler echocardiographic results. A n n Thorac Surg, 1998. 66(5): p. 1692-7. 107. Kshettry, V . R . and L . T . Kanda, Prospective study of mitral valve repair with the CarboMedics AnnuloFlex annuloplasty system: effectiveness and safety after one year. J Heart Valve Dis , 2005.14(1): p. 105-13. 108. Umana, J.P., et al., "Bow-tie" mitral valve repair: an adjuvant technique for ischemic mitral regurgitation. A n n Thorac Surg, 1998. 66(5): p. 1640-6. 109. Glasson, J.R., et al., Mitral annular size and shape in sheep with annuloplasty rings. J Thorac Cardiovasc Surg, 1999. 117(2): p. 302-9. 110. Dall 'Agata, A . , et al., Cosgrove-Edwards mitral ring dynamics measured with transesophageal three-dimensional echocardiography. A n n Thorac Surg, 1998. 65(2): p. 485-90. 111. Ibrahim, M . F . and T .E . David, Mitral stenosis after mitral valve repair for non-rheumatic mitral regurgitation. A n n Thorac Surg, 2002. 73(1): p. 34-6. 112. Parish, L . M . , et al., The dynamic anterior mitral annulus. A n n Thorac Surg, 2004. 78(4): p. 1248-55. 93 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092755/manifest

Comment

Related Items